Bulahdelah Flood Management Study

REVIEW OF BULAHDELAH FLOODPLAIN RISK MANAGEMENT STUDY AND PLAN FLOOD STUDY REVIEW

Report MHL2789

July 2023

Prepared for: MidCoast Council

Cover Photograph: Flooding of Stroud Street, Looking North, on 8 June 2007. Courtesy of Eric Saville

Review of Bulahdelah floodplain risk management study and plan

Flood study review

Report MHL2789

July 2023

Bronson McPherson Director of Engineering, MHL

110b King Street

Manly Vale NSW 2093

T: 02 9949 0200

E: Bronson.McPherson@mhl.nsw.gov.au

W: www.mhl.nsw.gov.au

Document Control

COPYRIGHT NOTICE

This document, Review of Bulahdelah Floodplain Risk Study and Plan 2023, is licensed under the Creative Commons Attribution 4.0 Licence, unless otherwise indicated. Please give attribution to: © MidCoast Council 2023

We also request that you observe and retain any notices that may accompany this material as part of the attribution.

Notice Identifying Other Material and/or Rights in this Publication:

The author of this document has taken steps to both identify third-party material and secure permission for its reproduction and reuse. However, please note that where these third-party materials are not licensed under a Creative Commons licence, or similar terms of use, you should obtain permission from the rights holder to reuse their material beyond the ways you are permitted to use them under the Copyright Act 1968 Please see the Table of References at the rear of this document for a list identifying other material and/or rights in this document.

Further Information

For further information about the copyright in this document, please contact: MidCoast Council

Yalawanyi Ganya 2 Biripi Way PO Box 482, Taree NSW 2430 council@midcoast.nsw.gov.au

02 7955 7777

DISCLAIMER

The Creative Commons Attribution 4.0 Licence contains a Disclaimer of Warranties and Limitation of Liability. In addition: This document (and its associated data or other collateral materials, if any, collectively referred to herein as the ‘document’) were produced by Manly Hydraulics Laboratory for MidCoast Council only. The views expressed in the document are those of the author(s), or as informed by the FRM Committee, and do not necessarily represent the views of the MidCoast Council. Reuse of this study or its associated data by anyone for any other purpose could result in error and/or loss. You should obtain professional advice before making decisions based upon the contents of this document.

Report Classification

☒ Public Report existence and contents publicly available.

Report No. MHL2789

First published as draft in December 2020

SEO SYSTEM Manly Hydraulics Laboratory is Safety, Environment & Quality System Certified to AS/NZS 4801, ISO 14001 and ISO 9001

110B King Street

Manly Vale NSW 2093

T 02 9949 0200

TTY 1300 301 181

ABN 20 770 707 468 www.mhl.nsw.gov.au

Foreword

NSW government’s professional specialist advisor, Manly Hydraulics Laboratory (MHL) were commissioned by MidCoast Council to undertake a review of the Bulahdelah Floodplain Risk Management Study and Plan.

The report was prepared by Armaghan Severi, Matthieu Glatz and Bronson McPherson. This report is release by consent report classification and is for the use of MidCoast Council only for the purposes of the specific engagement of services agreed to be provided by MHL. No part of it may, in any form or by any means, be reproduced, altered, manipulated, stored in a retrieval system or transmitted without prior written consent without prior written permission from the Director, Manly Hydraulics Laboratory.

© Crown 2021

Classification: Release by consent

Executive summary

The Review of Bulahdelah Floodplain Risk Management Study and Plan (FRMSP) consists of a two-stage review: the first stage consists of reviewing the flood study and the second stage consists of reviewing the floodplain risk management study and plan. This report documents the review of the flood study and has been prepared in accordance with the New South Wales Government’s Floodplain Development Manual (2005) A flood study is the first step of the floodplain management process set up to reduce flood risks and private/public losses resulting from flood while using eco-friendly solution where possible.

Manly Hydraulic Laboratory (MHL) were engaged by MidCoast Council (Council) to undertake the Review of Bulahdelah Floodplain Risk Management Study and Plan The original flood study was completed in 1991 using a 1D model and the FRMSP was completed in 2002-2003. The previous studies, while providing relevant information that relates to the flooding of Bulahdelah, were completed prior to the creation of the latest guidelines It was therefore required to update the model of the area using a two-dimensional TUFLOW model with inflow from an updated WBNM hydrological model for the estimation of mainstream (i.e. from the Myall and Crawford Rivers) and potential local overland (i.e. in Bulahdelah township) flood behaviour in the study area. This also took into account the performance of the stormwater drainage network including overflows from the drainage network.

The study outputs will inform the subsequent review of the FRMSP as well as decision making for investing in the floodplain; managing flood risk through prevention, preparedness, response and recovery activities; pricing insurance, and informing and educating the community on flood risk and response to floods. Each of these areas has different user groups or stakeholder, whose needs vary. Meeting the requirements of the identified end user groups and stakeholders in regards to flooding, is a key objective of this study.

The study has been overseen and guided by the Floodplain Management Advisory Committee of MidCoast Council, which includes representatives from key stakeholder and end user groups.

The flood study review has been completed to provide a detailed flooding assessment of Bulahdelah and the surrounding local catchment. The objective of this study is to improve understanding of flood behaviour and impacts, and better inform management of flood risk in the study area. The study also provides a sound technical basis for further flood risk management investigation in the area.

The key components of the flooding assessment included:

• Review of previous studies and available data

• Community consultation

• Hydrological analysis and modelling

• Hydraulic analysis and modelling

• Calibration and validation of the hydraulic model

• Sensitivity analysis

• Flood mapping

• Description of consequences of flooding

• Development of a draft flood study review report followed by a final report

The hydraulic model was calibrated and validated against recent historical events including the March 2021 flood event. The model provided a fair match with available data such as recorded water levels at local gauges, historical photographs information provided by local community members and flood marks measured around Bulahdelah by MHL’s project team following the March 2021 event

The flood maps appended to this report are presenting the flood levels, depths and velocities for the critical duration and rainfall pattern of a full set of events including the 20%, 5%, 1%, 1 in 200, 1 in 500 AEP and Probable Maximum Flood (PMF) events and represent an envelope of the critical duration/pattern of a selected representative catchment for the Bulahdelah township and the critical duration/pattern for the Myall River. The township catchment is very flashy with very short critical durations between 45 minutes and 2 hours to reach the peak level while the river catchment have typical critical durations ranging between 6 hours and 18 hours

Sensitivity analysis highlighted the following points:

• Tailwater conditions (including sea level rise) typically have relatively small impact on river flooding levels due to a number of natural choke points located at various locations along the Myall River raising the level much higher than the downstream level However, for smaller catchment events that do not lead to a major response in the overall river catchment flooding, changes in Bombah Broadwater levels will propagate directly up the Myall River.

• Changes in roughness or in antecedent conditions of the catchment (wet/dry catchment leading to varying losses) could have moderate impacts on river flooding and low impact on overland flooding.

• Blockages of structures can have severe impact in areas with no gravity flow that only rely on the drainage network (e.g. ponding area), and maintaining the pits and pipes network is essential to avoid exacerbating the flooding in such locations. It is also noted that no blockage can lead to small increases in flood level at the downstream end of the township.

The above results allowed the definition of the flood hazard (i.e. H1-H6 flood life hazard categories) and hydraulic categories in the Bulahdelah catchment. These have been created and mapped to inform development control planning.

Results of the model allow the identification of main flooding areas, key infrastructure impacted by flooding and road closures around the catchments. Key infrastructure typically may have impeded access during severe flood events rather than being flooded, except during the PMF event.

Key issue areas potentially subject to deep flooding from the river include:

• Properties located west of Stroud Street and south of Jackson Street (including most properties along Jackson Street);

• Properties located in the two blocks enclosed within Prince Street, Richmond Street,

Alexandra Street and Jackson Street;

• Properties east of Stroud Street and west of Crawford Street between Meade Street and Jackson Street; and

• The area around the intersection of River Street and Richmond Street.

Key problem areas subject to shallower overland flooding include:

• Properties between Myall Street and Edgar Street at the southern end of the streets;

• Properties located along the drainage channel linking Stroud Street to Richmond Street via Alexandra Street;

• Properties along the flowpath linking Mackenzie Street to Crawford Street via Harold Street, Rosewood Street, Bulahdelah Way and Meade Street;

• Properties along the flowpath linking Bulahdelah Way to Crawford Street halfway between Stuart Street and Blanch Street, as well as properties along the western side of Crawford Street between this flowpath and Stuart Street; and

• Properties at the northeastern end of Suncrest Close.

Stroud Street, Jackson Street and Markwell Road near Markwell Bridge are inundated during a 20% AEP. Apart from Stroud Street, road closures predominantly occur on secondary roads with most of the major road closures occurring for the PMF only. It is also noted that given the flashy behaviour of the catchment, flooding and road closures in the upper township catchment would be of relatively short duration, while flooding of the areas surrounding the river may last from several hours to a few days

1 Introduction

NSW Government’s Manly Hydraulics Laboratory (MHL) was engaged by MidCoast Council (Council) to undertake a review of the Floodplain Risk Management Study and Plan (FRMSP) for Bulahdelah following the NSW Government’s Flood Prone Land Policy update and with financial support from the Biodiversity, Conservation and Science group (BCS) of the NSW Department of Planning Industry and Environment (DPIE), as well as technical support from DPIE and the NSW State Emergency Services

There have been various hydrologic and hydraulic studies undertaken for the Bulahdelah catchments. The most recent study was the Myall River Floodplain Risk Management Study and Plan (DIPNR, 2003). The 2003 FRMSP reported the flood behaviour across the catchment for a range of design flood events for existing topographic and development conditions.

The objective of the present study is to review the 2003 Myall River Floodplain Risk Management Study and Plan for Bulahdelah focusing on flooding from Myall River and its tributaries as local catchment flows have already been studied in the Bulahdelah Stormwater Management Plan (Cardno, 2020)

2 Background

2.1 Study area

The focusof the present investigation isthe township of Bulahdelah and Frys Creekcatchment. The township of Bulahdelah is located approximately 70 km north-east of Newcastle, between the Pacific Highway and the Myall River near its confluence with the Crawford River.

The Myall River catchment upstream of Bulahdelah has a catchment area of approximately 370 km2 including its tributaries catchment areas of 125 km2 for the Crawford River and 18 km2 for Frys Creek. The catchment is steep upstream of Bulahdelah and flattens downstream of the township.

It is understood that the dominant flooding mechanism is the mainstream flooding from the Myall River and the Crawford River as no structures are protecting the township from floods. However, it is noted that the township can also be subject to local overland flooding. The flood behaviour has also been influenced by the construction of the new Pacific Highway in 2013. The study area is presented in Figure 2.1.

2.2 History of flooding and rainfall

Several significant historical floods were reported in 1897 and 1927. Smaller floods occurred in 1947 and 1953, and less severe floods happened in 1985 and 1987

The largest flood at Bulahdelah was reported in 1897, reaching the level of 5.5 m AHD on the upstream side of the highway bridge (at the existing Myall River Bridge) with an estimated magnitude of 100 year Average Recurrence Interval / ARI (equivalent to 1% Annual Exceedance Probability / AEP) (Public Works, 1991) The next largest reported flood occurred in 1927 reaching a levelof 5.1 m AHD at the highway bridge (at the existing Myall River Bridge) with an estimated magnitude of 50 year ARI (2% AEP). A flood level of 6.4 m AHD was reported at the confluence of the Myall and Crawford Rivers approximately 500 m upstream. A flood level of 6.5 m AHD was reported near the water supply weir in the Crawford River for both the 1897 and 1927 floods (Public Works, 1991)

A peak flood level of 3.7 m was reported at the highway bridge in October 1985 and November 1987. However, the 1985 flood was higher upstream of the bridge (Public Works, 1991) The largest events recorded by the water level gauge in Bulahdelah (installed in November 1984) are March 2021, April 2015, March 2013 and January 2016 events with the corresponding peak flood level of 4.3 m, 4.1 m, 3.9 m and 3.8 m The March 2021 flood occurred during the development of this study and MHL’s project team undertook some post-event surveys as described in Section 4.8

Some photographs of the past events have been provided by community members. The locations and orientation of the photographs are provided in Figure 2.2, and some examples are provided in Figure 2.3 to Figure 2.7.

Historical newspaper articles describing flooding at Bulahdelah have been provided in Figure 2 8 and Figure 2 9

Photograph providedby community

Legend

Cadastre

Photographlocationand directionduringflood eventon: 8June2007 21April2015

ReportMHL2789

ReviewofBulahdelah FloodplainRisk ManagementStudy&Plan

Figure2.2

JacksonSt

Figure2.5

Figure2.4

Figure2.3

Figure2.6

Figure2.7

Figure 2 3 Stroud Street, Looking South, 8 June 2007 (courtesy of Eric Saville)

Figure 2.4 Stroud Street, Looking North, 8 June 2007 (courtesy of Eric Saville)

Figure 2.5 Jackson Street, Looking East, 8 June 2007 (courtesy of Eric Saville)

Figure 2.6 Jackson Street, Looking West, 21 April 2015 (courtesy of Eric Saville)

Figure 2.7 Stroud Street, Looking South, 21 April 2015 (courtesy of Eric Saville)

Figure 2.8 Extract of a newspaper article, Newcastle Sun Archives from Wednesday 20 April 1927 (https://trove.nla.gov.au/newspaper/article/125142777)

2.3 Relevant policies, legislation and guidance

2.3.1 National provisions

2.3.1.1 AustralianRainfallandRunoff,2019

Australian Rainfall and Runoff (ARR) is a national guideline document, data and software suite that is used for the estimation of design flood characteristics in Australia. This is the 4th edition of ARR after the 1st edition was released by Engineers Australia in 1958. Geoscience Australia supports ARR as part of its role to provide authoritative, independent information and advice to the Australian Government and other stakeholders to support risk mitigation and community resilience.

ARR is pivotal to the safety and sustainability of Australian infrastructure, communities and the environment. It is an important component in the provision of reliable and robust estimates of flood risk. Consistent use of ARR ensures that development does not occur in high-risk areas and that infrastructure is appropriately designed.

2.3.1.2 BuildingCodeofAustralia

The 2016 edition of the Building Code of Australia (BCA) introduced new requirements related

Figure 2.9 Extract of a newspaper article, Newcastle Sun Archives from Tuesday 26 March 1946 (https://trove.nla.gov.au/newspaper/article/158272189)

to building in Flood Hazard Areas (FHAs), which provide a minimum construction standard across Australia for specified building classifications in FHAs up to the Defined Flood Event (DFE).

The DFE is analogous to the planning flood event and is most commonly the 1% AEP flood. FHAs are defined in the BCA as encompassing land lower than the flood hazard level (FHL), which in turn is defined as ‘the flood level used to determine the height of floors in a building and represents the DFE plus the ‘freeboard’. Therefore, FHAs would typically be defined as those areas falling within the flood planning area.

Volume One, BP1.4, and Volume Two, P2.1.2 specify the Performance Requirements for the construction of buildings in FHAs. They only apply to buildings or parts of buildings of Classes 1, 2, 3, 4 (residential), 9a (health-care) and 9c (aged-care). These Performance Requirements require a building in an FHA to be designed and constructed to resist flotation, collapse and significant permanent movement resulting from flood actions during the DFE. The actions and requirements to be considered to satisfy this performance requirement include but are not limited to:

• Flood actions

• Elevation requirements

• Foundation and footing requirements

• Requirements for enclosures below the flood hazard level

• Requirements for structural connections

• Material requirements

• Requirements for utilities

• Requirements for occupant egress

The Deemed-to-Satisfy (DTS) provisions of Volume One, B1.6, and Volume Two, 3.10.3.0 require buildings in the classes described above and located in FHAs to comply with the Australian Building Code Board (ABCB)Standard for Construction of Buildings in Flood Hazard Areas 2012 (the ABCB Standard).

The ABCB Standard specifies detailed requirements for the construction of buildings to which the BCA requirements apply, including:

• Resistance in the DFE to flood actions including hydrostatic actions, hydrodynamic actions, debris actions, wave action and erosion and scour

• Floor height requirements, for example that the finished floor level of habitable rooms must be above the FHL

• The design of footing systems to prevent flotation, collapse or significant permanent movement

• The provision in any enclosures or openings to allow for automatic entry and exit of floodwater for all floods up to the FHL

• Ensuring that any attachments to the building are structurally adequate and do not reduce the structural capacity of the building during the DFE

• The use of flood-compatible structural materials below the FHL

• The siting of electrical switches above the FHL, and flood proofing of electrical conduits and cables installed below the FHL

• The design ofbalconiesetc. to allowa person in the building to be rescued byemergency services personnel, if rescue during a flood event up to the DFE is required

Building Circular BS13-004 (NSW Department of Planning and Infrastructure, 2013) summarises the scope of the BCA and how it relates to NSW planning arrangements. The scope of the ABCB Standard does not include parts of FHA that are subject to flow velocities exceeding 1.5 m/s or are subject to mudslide or landslide during periods of rainfall and runoff or are subject to storm surge or coastal wave action.

It is particularly noted that the Standard applies only up to the DFE, which typically will correspond to the level of the 1% AEP flood plus 0.5 m freeboard. The Building Circular emphasises that because of the possibility of rarer floods, the BCA provisions do not fully mitigate the risk to life from flooding.

The ABCB has also prepared an Information Handbook for the Construction of Buildings in Flood Hazard Areas. This Handbook provides additional information relating to the construction of buildings in FHA but is not mandatory or regulatory in nature.

In the NSW planning system, the BCA takes on importance for complying development under the State Environmental Planning Policy (Exempt and Complying Development Codes) 2008.

2.3.2 State provisions

2.3.2.1 EnvironmentalPlanningandAssessmentAct,1979 General

The NSW Environmental Planning and Assessment Act 1979 (EP&A Act) creates the mechanism for development assessment and determination by providing a legislative framework for development and protection of the environment from adverse impacts arising from development. The EP&A Act outlines the level of assessment required under State, regional and local planning legislation and identifies the responsible assessing authority.

Prior to development taking place in NSW a formal assessment and determination must be made of the proposed activity to ensure it complies with relevant planning controls and, according to its nature and scale, conforms with the principles of environmentally sustainable development.

Section 7.11 Development Contributions

Section 7.11 (previously Section 94) of the EP&A Act enables councils to collect contributions from developers for the provision of infrastructure that is necessary as a consequence of development. This can include roads, drainage, open space and community facilities. Each council must develop a Section 94 Contributions Plan which demonstrates a quantifiable link between the development intensification and the need for the additional infrastructure as well as a detailed costing of such infrastructure and formulae to be used to determine contributions from each type of development.

Section 9.1 Directions – Direction No. 4.3 (Flood Prone Land)

NSW flood-related planning requirements for local councils are set out in Ministerial Direction No. 4.3 Flood Prone Land, issued in 2007 under the then Section 117 (now Section 9.1) of the EP&A Act. It requires councils to ensure that development of flood prone land is consistent with the NSW Government’sFlood Prone Land Policy as set out in the Floodplain Development Manual (NSW Government, 2005). It requires provisions in a Local Environmental Plan on flood prone land to be commensurate with the flood hazard of that land. In particular, a planning proposal must not contain provisions that:

• Permit development in floodway areas

• Permit development that will result in significant flood impacts to other properties

• Permit a significant increase in the development of that land

• Are likely to result in a substantially increased requirement for government spending on flood mitigation measures, infrastructure or services

• Permit development to be carried out without development consent except for the purposes of agriculture, roads or exempt development

The Direction also requires that councils must not impose flood related development controls above the residential flood planning level (FPL, typically the 1% AEP flood plus 0.5m freeboard) for residential development on land, unless a relevant planning authority provides ‘adequate justification’ for those controls to the satisfaction of the Director-General. However, it is noted that some special flood considerations LEP clause now allows for flood planning levels above 1% AEP + 0.5 m.

Section 10.7 Planning Certificates

Planning certificates are a means of disclosing information about a parcel of land. Two types of information are provided in planning certificates: information under Section 10.7(2) and information under Section 10.7(5) of the EP&A Act. (Note that previously this clause was Section 149).

A planning certificate under Section 10.7(2) discloses matters relating to the land, including whether or not the land is affected by a policy that restricts the development of land. Those policies can be based on identified hazard risks (Environmental Planning and Assessment Regulation 2000, Clause 279 and Schedule 4 Clause 7), and whether development on the land is subject to flood-related development controls (EP&A Regulation, Schedule 4 Clause 7A). If no flood-related development controls apply to the land (such as for residential development in so-called ‘low’ risk areas above the FPL, unless ‘adequate justification’ has been satisfied), information describing the flood affectation of the land would not be indicated under Section 10.7(2). A lot that is a ‘flood control lot’ under the Codes SEPP is a prescribed matter for the purpose of a certificate under section 10.7(2).

A planning certificate may also include information under Section 10.7(5). This allows a council to provide advice on other relevant matters affecting land. This can include past, current or future issues.

Inclusion of a planning certificate containing information prescribed under section 10.7(2) is a mandatory part of the property conveyancing process in NSW. The conveyancing process

does not mandate the inclusion of information under section 10.7(5) but any purchaser may request such information be provided, pending payment of a fee to the issuing council.

2.3.2.2

StateEnvironmentalPlanningPolicies(SEPPs)

SEPPs are the highest level of planning instrument and generally prevail over Local Environmental Plans.

SEPP (Housing for Seniors or People with a Disability) 2004

State Environmental Planning Policy (Housing for Seniors or People with a Disability) 2004 aims to encourage the provision of housing (including residential care facilities) that will increase the supply of residences that meet the needs of seniors or people with a disability. This is achieved by setting aside local planning controls that would prevent such development.

Clause 4(6) and Schedule 1 indicate that the policy does not apply to land identified in another environmental planning instrument as being, amongst other descriptors, a floodway or high flooding hazard.

SEPP

(Infrastructure) 2007

State Environmental Planning Policy (Infrastructure) 2007 aims to facilitate the effective delivery of infrastructure across the State by identifying development permissible without consent. SEPP (Infrastructure) 2007 allows Council to undertake stormwater and flood mitigation work without development consent.

SEPP (Exempt and

Complying Development Codes) 2008

A very important SEPP is State Environmental Planning Policy (Exempt and Complying Development Codes) 2008, which defines development which is exempt from obtaining development consent and other development which does not require development consent if it complies with certain criteria.

Clause 1.5 of this ‘Codes’ SEPP defines a ‘flood control lot’ as ‘a lot to which flood related development controls apply in respect of development for the purposes of dwelling houses, dual occupancies, multi dwelling housing or residential flat buildings (other than development for the purposes of group homes or seniors housing)’. These development controls may apply through a Local Environmental Plan (LEP) or DCP. Exempt development is not permitted on flood control lots but some complying development is permitted.

Clause 3.5 states that complying development is permitted on flood control lots where a Council or professional engineer can certify that the part of the lot proposed for development is not a flood storage area, floodway area, flow path, high hazard area or high-risk area. The Codes SEPP specifies various controls in relation to floor levels, flood compatible materials, structural stability (up to the PMF if on-site refuge is proposed), flood affectation, safe evacuation, car parking and driveways.

In addition, Clause 1.18(1)(c) of the Codes SEPP indicates that complying development must meet the relevant provisions of the Building Code of Australia.

SEPP

(Coastal Management) 2018

SEPP (Coastal Management) 2018 aims to promote an integrated and co-ordinated approach to land use planning in the coastal zone. For areas mapped as ‘coastal wetland and littoral

rainforests’ – including sizeable areas in the study area near the three lakes – development consent is required for the clearing of native vegetation, and for earthworks, construction of a levee, draining the land and environmental protection works, and for any other development. For areas mapped as ‘coastal environment areas’ – covering much of the study area –development consent must not be granted unless the consent authority has considered whether the proposed development is likely to cause an adverse impact on “the integrity and resilience of the biophysical, hydrological (surface and groundwater) and ecological environment” amongst other factors. The development must be designed, sited and managed to either avoid, minimise or mitigate adverse impacts.

2.3.2.3 NSWFloodRelatedManuals Floodplain Development Manual, 2005

The Floodplain Development Manual 2005 (the Manual) was gazetted on 6 May 2005 and relates to the development of flood liable land. It incorporates the NSW Flood Prone Land Policy, which aims to reduce the impacts of flooding and flood liability on individual owners and occupiers of flood prone property and to reduce private and public losses resulting from floods, using ecologically positive methods wherever possible. To implement this policy and achieve these objectives, the Manual espouses a merit approach for development decisions in the floodplain, taking into account social, economic, ecological and flooding considerations. The Manual confirms that responsibility for management of flood risk remains with local government. It assists councils in their management ofthe use and development of flood prone land by providing guidance in the development and implementation of local floodplain risk management plans.

At the time of preparing this report, the Floodplain Development Manual is being updated.

Guideline on Development Controls on Low Flood Risk Areas, 2007

The Guideline on Development Controls on Low Flood Risk Areas – Floodplain Development Manual (the Guideline) was issued on 31 January 2007 as part of Planning Circular PS 07003 at the same time as the Section 117 (now Section 9.1) Directive described previously. The Guideline is intended to be read as part of the Floodplain Development Manual

It stipulates that unless there are exceptional circumstances, councils should adopt the 100year flood as the flood planning level (FPL) for residential development and that ‘unless there are exceptional circumstances, councils should not impose flood related development controls on residential development on land … that is above the residential FPL’.

An adequate freeboard should then be applied to the 100-year flood level to allow for potential inaccuracies in available data and limitation of the flood models.

Flood related development controls are not defined but would include any development standards relating to flooding applying to land, thatare a matter for consideration under Section 4.15 (previously Section 79C) of the EP&A Act.

The Guideline states that councils should not include a notation for residential development on Section 10.7 (previously Section 149) certificates for land above the residential FPL if no flood related development controls apply to the land. However, the Guideline does include the reminder that councils can include ‘such other relevant factors affecting the land that the council may be aware [of]’ under Section 10.7(5) of the EP&A Act.

In proposing a case for exceptional circumstances, a council would need to demonstrate that a different FPL was required for the management of residential development due to local flood behaviour, flood history, associated flood hazards or a particular historic flood. Justification for exceptional circumstances would need to be agreed by relevant State Government departments prior to exhibition of a draft local environmental plan or a draft development control plan that proposes to introduce flood related development controls on residential development above the default FPL.

At the time of preparing this report, the Guideline is being reviewed.

2.3.3 Local provision

In NSW, local government councils are responsible for managing flood risk within their Local Government Areas (LGAs). An LEP is used to establish what land uses are permissible and/or prohibited on land within the LGA and sets out high level flood planning objectives and requirements. A Development Control Plan (DCP) sets the standards, controls and regulations that apply when carrying out development or building work on land. This section briefly describes and reviews the flood-related controls within the MidCoast policies, with a view to flood behaviour in the Bulahdelah study area.

2.4 Land use

Land-use zoning in Bulahdelah is defined in the Great Lakes Local Environmental Plan (LEP) 2014, shown in Figure 2.10 The majority of the township itself is zoned as “RU5 Village” with a small area to the southwest and an area in the middle of the township zoned as “RE1 Public Recreation”. The Pacific Highway which connectsBulahdelah to neighbouring towns is located to the southeast of the town and zoned as “SP2 Infrastructure”. Figure 2.10 shows that an area zoned as “C1 National Parks and Nature Reserves” is located to the north of the town. The majorityof the floodplain west and north of Bulahdelah is zoned as “RU2 RuralLandscape” and downstream of Bulahdelah as “C2 – Environmental Conservation” or “C3 – Environmental Management”. The vegetated and elevated areas around the ridges of the catchment and Bombah Broadwater are zoned as either “C1 – National Parks and Nature Reserves” or “RU3 – Forestry”. Myall River itself is zoned as a “W2 Recreational Waterways” along and upstream of Bulahdelah and as “W1 – Natural Waterways” downstream of the township

2.5 Demographic overview

Understanding the social characteristics of the study area can help ensure appropriate risk management practices are adopted and shape the methods used for community engagement. House tenure and age distribution data obtained from census data can indicate the community’sexperience with recent flood events,and hence an indication of community’s flood awareness. As per The Bureau of Meteorology Flood Preparedness Manual, using the population census data and other information held by councils and state agencies can help to identify the potential number and location of people in an area with special needs or requiring additional support during floods (Australian Government (Attorney – General’s Department), 2009). The relevant information has been extracted from the 2016 Census for the town of Bulahdelah (and surrounds) and tabulated in Table 2.1

Classification: Release by consent

Landuse

Legend

Studyarea

LandZoningCategory

C1-NationalParksand NatureReserves

C2-Environmental Conservation

C3-Environmental Management

R2-LowDensity Residential

R5-LargeLot

Residential

RE1-PublicRecreation

RU2-RuralLandscape

RU3-Forestry

RU5-Village

SP2-Infrastructure

W1-NaturalWaterways

W2-Recreational Waterways

ReportMHL2789

ReviewofBulahdelah FloodplainRisk ManagementStudyandPlan

Source: https://www.abs.gov.au/census/find-census-data/quickstats/2021/SAL10646

466 (75.2%, either outright or with a mortgage)

Language English only is spoken at home: 1393 (90.6%) A non-English language is spoken at home: 29 (4 7%)

Classification: Release by consent

Table 2.1 Bulahdelah demographic overview based on the 2021 census. Bulahdelah Demographic Overview

3 Previous studies

3.1 Bulahdelah Flood Appraisal, NSW Public Works Department, 1991

This report was prepared for Great Lakes Shire Council, by the NSW Public Works Department (PWD) in October 1991 to assess flood behaviour for the Myall River floodplain in the vicinity of Bulahdelah (Public Works, 1991). The study area extended along the Myall River from 0.43 km downstream of the Pacific Highway bridge to about 2 km upstream of the bridge at Lee Street.

The study utilised the WBNM model with eight sub-catchments to produce the discharge hydrographs for the October 1985 and November 1987 flood events as well as 1%, 2% and 5% AEP events. WBNM hydrologic model was set using the lag parameter of C = 1.29 A onedimensional hydraulic model of MIKE 11 was established to study the flood behaviour across the study area. The Bulahdelah MIKE 11 model consisted of seven surveyed cross-sections (surveyed by the PWD in 1987 and 1988) and the Pacific Highwaybridge profile. The upstream and downstream boundaries were defined as discharge hydrographs at the confluence of the Crawford River and a stage hydrograph, respectively.

The study utilised daily rainfall and pluviography data to describe the areal and temporal distribution of rainfall No pluviograph station was located within the Myall River catchment therefore, the closest ones acquired for the study including Upper Johnson Creek, Nabiac, Maryville (Newcastle) Williamtown and Taree. Long term daily rainfall data were obtained from BoM station located at Bulahdelah Post Office and several stations located around the boundary of the catchment Further daily rainfall records were used to undertake the Flood Study from unofficial gauges located at Upper Myall (Tank Creek), Markwell (Deep Creek) and Rosenthal (Gloucester Road) No adequate details of the unofficial rainfall gauges are available. ARR 1987 methodology was used to determine the critical duration of design storm events, resulting in the critical duration of 36 hours for the 1% AEP event.

The study documented that the most information on flooding was available for 1985 and 1987 events (for further information on historical flood events see Section 2.2). Therefore, the hydraulic model was calibrated and verified using the October 1985 and November 1987 floods, respectively.

For the 1985 flood, an initial loss of 0 mm and a continuing loss of 2.5 mm/hr were adopted due to considerable rainfall prior to the event. An initial rainfall loss of 21 mm and continuing losses of 2.5 mm/hr were adopted for the 1987 flood and the design floods.

Sensitivity analysis was conducted to investigate the impacts of the tailwater conditions at the Muirs Creek confluence on the hydraulic model outcomes. The 1% AEP flood event hydrograph at the Highway Bridge was insensitive to the range of tailwater conditions. The 5% AEP flood event hydrograph at the Highway Bridge was relatively insensitive to the range of tailwater conditions, with 200 mm variation within the range of tested tailwater conditions. A single tailwater curve with a peak level of 3 m AHD was adopted for modelling the design events.

The key finds from the report are summarised below:

• A constant Manning’s roughness coefficient of 0.037 for the main channel and 0.05 for the floodplain resulted in the best outcomes.

• In 1% AEP event, a critical storm duration of 36 hours was found to produce a peak discharge of 2100, 750 and 1400 m3/s at the Bulahdelah bridge, Crawford River and upstream of Crawford River, respectively.

• It was documented that the 1897 flood event is the largest flood on record with a flood level of RL 5.5 m reported in the vicinity of the Highway Bridge.

• The simulations of 1% AEP event resulted in a flood level of approximately 50 mm higher than the 1897 event in the vicinity of the highway Bridge.

• Also, it was reported that the 1927 flood event is the second largest flood on record with a flood level of RL 5.0 m.

• The calculated flood level of a 2% AEP event resulted in a flood level of approximately 100 mm higher than the 1927 record.

The review of the Bulahdelah Flood Appraisal revealed that the WBNM and MIKE11 models utilised in the study are very limited in terms of the number of sub-catchments to represent the study area and cross-sections to represent the 1D model. Therefore, there is a need for establishing hydrologic and hydraulic models with adequate details to appropriately represent the study area

3.2 Frys Creek Flood Study, NSW Public Works Department, 1994

This report was prepared for the Great Lakes Shire Council, by the NSW Public Works Department in July 1994 to assess flood behaviour in the lower reaches of the Frys Creek (Public Works, 1994)

The study adopted and adapted the hydrological WBNM model and the one-dimensional hydraulic model of MIKE 11 developed for the completion of Bulahdelah Flood Appraisal (1991). To meet the objectives of the study, the MIKE 11 model was extended to include Frys Creek. Figure 3 1 showed the model schematization including the branches, weirs and culverts. The hydraulic model was calibrated using the recorded flood level in November 1987 event near Lee Street, Frys Creek Bridge and Markwell Road Bridge The simulated peak water level, discharge and velocity results for the 1% and 5% AEP events are tabulated in Table 3.1 to Table 3.3.

Figure 3 1 Model schematization (NSW Public Works Department, 1994)

Table 3 1 Frys Creek Flood study summary of water level results

Table 3.2 Frys Creek Flood study summary of discharge results

Table 3.3 Frys Creek Flood study summary of velocity results

3.3 Myall River Floodplain Risk Management Study for Bulahdelah, Department of Land and Water Conservation, 2002

The Floodplain Risk Management Study Report was completed for the Great Lakes Shire Council, by the Department of Land and Water Conservation in August 2002 to assess the potential floodplain risk management measures to alleviate the flooding issues caused by mainstream flooding at Bulahdelah (Department of Land and Water Conservation, 2002) The study area covers the floodplains of the Myall River and its main tributaries including Myall River from the Markwell Road Bridge to downstream of the Pacific highway Bridge, Frys Creek from upstream of the Markwell Road Bridge to the Myall River and Crawford River from the confluence with Wild Cattle Creek to the Myall River. The study only investigated the mainstream flooding due to the Myall River and its main tributaries.

The key finds from the report are summarised below:

• The major flood events can have a significant impact on Bulahdelah including significant property damages and danger to personal safety

• A range of flood levels was observed from 5.6 m AHD to 6.1 m AHD from upstream of the Pacific Highway to Lee Street at Bulahdelah. The peak average mainstream flow velocity varied from 3.1 m/s in the vicinity of Lee Street to 1.3 m/s upstream of the confluence of the Crawford River. Peak average overbank velocities on the eastern side of the Myall River is ranging from 0.4 to 0.8 m/s. Flood levels for the 1% and 2% AEP events were similar to the 1897 and 1927 flood levels in this vicinity

• In an extreme event, the flood level is generally in the order of 1.8 m higher than the 1% AEP event.

• In the 1% AEP event, around 45 residential and 7 non-residential properties are subject to inundation above the floor level resulting in the potential flood damage of $1,700,000.

• In an extreme event, the number of residential properties subject to inundation above the floor level is likely to double.

• The main channels of Myall River, Crawford River and Frys Creek are categorised as floodway. The remaining floodplain area at Bulahdelah was categorised as flood storage and some areas as flood fringe.

• This study recommended a voluntary house raising and voluntary purchase scheme, updating the Council’s 1985 Flood Management Policy and associated development controls, installation of a flood warning system with associated emergency management planning, community education and awareness.

The study tabulated the simulated peak water level results for the 5%, 2%, 1% AEP and extreme events reported in Table 3.4 Also, peak flow for the Myall River and tributaries for the 5%, 2%, 1% AEP were presented in Table 3.5

by

Table 3.4 Summary of water level results.

Table 3.5 Peak flows for Myall River and tributaries.

3.4 Myall River Floodplain Risk Management Plan for Bulahdelah, Department of Infrastructure, Planning and Natural Resources, 2003

This report was prepared for the Great Lakes Shire Council, by the Department of Infrastructure, Planning and Natural Resources in August 2003 to assess the potential floodplain risk management measures to alleviate the flooding issues at Bulahdelah (Department of Infrastructure, Planning and Natural Resources, 2003). The flood management measures recommended in this plan tabulated in Table 3.6

Table 3.6 Proposed flood management measures.

1 (Undertaken immediately)

2a (Undertaken as soon as subsided funds are available)

2b

2c

Updating the development controls and S149 certificates

Flood response plan

Flood warning system

Flood education and awareness

2d Ongoing data collection

3 (Undertaken every 5 years) Review of the Floodplain Management Plan (2003).

4 Voluntary house raising of all below 2% AEP flood to 0.5 m above the peak flood level in 1% AEP event.

5 Voluntary purchase of all below 5% AEP flood level that can’t be raised

3.5 Bulahdelah Upgrading the Pacific Highway, Parsons Brinckerhoff, 2004

This report was prepared for the Road and Traffic Authority (RTA), by Parsons Brinckerhoff in November 2004 to assess the flood impact associated with the construction of the new bridge and embankments associated with the Bulahdelah Bypass due to the highway upgrade.

The study utilised the MIKE 11 model prepared for the Myall River Floodplain Risk Management Study for Bulahdelah by the department of Land and Water Conservation. The model was based on surveyed cross-sections and calibrated against 1987 flood event by adjusting the roughness coefficient. Roughness coefficient of 0.09 and 0.06 were applied in the overbanks and in the channel in the vicinity of the bends downstream of the proposed bridge. Moreover, a steady-state one-dimensional backwater analysis software HEC-RAS was used to assess the impact of the proposed structures crossing the Myall River and floodplain.

The

study investigated the impact of several bridge options on flood flows in the Myall River

The report documented the simulated water level for the adopted option including a 192 m bridge opening, approximately 204 m2 of culvert openings in the embankment of the proposed Upgrade and 23 m2 of culvert openings in the embankment of the existing highway

3.6 Lower Myall River and Myall Lakes Flood Study, BMT, 2015

This report was completed for the Great Lakes Council by BMT in June 2015 to define the existing flood behaviour in the lake and river and establish the basis for subsequent floodplain management measures (BMT, 2015).

The study documented that the lower Myall River catchment covers an area of approximately 900 km2 and is located north of Port Stephens in the Great Lakes Council Local Government Area (LGA). Myall River is the major contributor to the lake system which drains an area of approximately445 km2 and enters the lake system at the north-western side of the Broadwater. The Lake system is drained by the 28 km long Lower Myall River which exits the Broadwater at Tamboy and drains into Port Stephens at Corrie Island via Corrie Creek and Paddy Marrs Inlet.

The study utilised RAFTS-XP hydrologic model with 61 sub-catchments to produce the discharge hydrographs. A two-dimensional hydraulic model of TUFLOW was established to study the flood behaviour across the study area. The TUFLOW model was developed using a 20 m grid to represent the Lower Myall River channel and floodplain and a 50 m grid to represent the Lakes and Myall channel and floodplain. The upstream boundary was defined as flow versus time inputs at major sub-catchments inflow points and along the modelled watercourses. The downstream boundary was modelled corresponding to the tidal water level of Port Stephens.

Myall Lake system and Lower Myall River is flooded by a combination of fluvial and ocean storm events. The following historical floods and their associated peak water levels were identified by MHL (1993) and are based on PWD (1980):

• 1890s - 3.7 m AHD (Myall Lakes);

• 1927 - 2.7 to 3.2 m AHD (Myall Lakes);

• 1963 - 2.2 m AHD (Tamboy, Myall Lakes);

• March 1977 - 1.2 m AHD (Bombah Point, Myall Lakes);

• March 1978 - 1.31 m AHD (Bombah Point, Myall Lakes); and

• May 1978 - 1.3 m AHD (Bombah Point, Myall Lakes).

Inspection of the Bombah Point water level gauge from July 2001 to May 2012 indicates several other significant flood events as presented in Table 3 7

* This is an estimated value as discussed in Section 5.5.1.2 of Lower Myall River and Myall Lake Flood Study.

The model was calibrated against July 2011 flood event and validated against May 2003 and April 2008 flood events. During the Flood Study, a significant flood event (in early March 2013) occurred and was used as a validation event. In addition to the four flood events, a tidal calibration exercise was undertaken using data collected by MHL in September 2009. An initial rainfall loss of 15 mm and continuing losses of 0.5 mm/hr were adopted for May 2003, April 2008 and July 2011 floods and the design floods. For the 2013 flood, an initial loss of 0 mm and a continuing loss of 0.1 mm/hr were adopted due to considerable rainfall on the 23rd February 2013.

Lake water level at the start of the July 2011 event is approximately 0.5 m AHD with a peak lake level of 1.27 m AHD occurring on the 25th July 2011. Lake water level at the start of the event on the 25th May is 0.78 m AHD with a peak lake level of 1.4 m AHD occurring on the 29th May. Lake level at the start of the event on the 18th April 2008 is 0.25 m AHD with a peak lake level of approximately 1.45 m AHD occurring on the 26th April 2008. Lake water level at the start of the event is approximately 0.8 m AHD with a peak lake level of 1.75 m AHD occurring on the 4th March 2013. The Bulahdelah water level gauge on the Myall River just downstream of the Pacific Highway Bridge is able to provide information on catchment response and at this location, the river water level is responsive to rainfall intensity, though is also affected by a tailwater from the Lakes.

Initial water levels in the Myall Lakes system for design flood events have been derived based on an analysis of water level statistics at Bombah Point between July 2001 and April 2012. An initial lake level of 0.5 m AHD was selected (for both catchment and ocean events) in consultation with OEH and GLC and matches the lake level exceeded only 10% of the time. A range of design flood conditions including the 50%, 20%, 10%, 5%, 2%, 1%, 0.5% AEP events and an extreme event (similar to PMF) were modelled and the results of simulated peak flood levels at several locations were documented Table 3.8 and Table 3.9 presented the peak flood levels at Myall Lake and Bombah Broadwater for three scenarios including catchment derived design events, ocean derived design events and combined conditions

Classification: Release by consent

Table 3.7 Peak flood levels at Bombah Points (m AHD).

The key finds from the report are summarised below:

• Results indicate that peak flood levels upstream of Monkey Jacket are attributed to catchment derived design flood events, while ocean derived flood events are more significant downstream of Monkey Jacket.

• Coincident ocean and catchment flood events cause a negligible increase in peak levels within the lake; however, depending on the timing of peak tides may increase peak flood levels below Kangaroo Island by up to 0.2 m.

• Impacts of climate change (including increases in rainfall intensity and sea-level rise) on flood risk along the Lower Myall were investigated. The outcomes revealed that an increase in mean sea level due to sea-level rise may pose a significant flood risk problem along the Tea Gardens peninsula, where ground elevations of only 1.5 m AHD are common. Low lying areas in Hawks Nest (with ground elevations below 2.0 m AHD) will also be affected by predicted SLR.

3.7 Flood Emergency Response Plan Bulahdelah Caravan Park, Cardno, 2017

This report was prepared by Cardno in April 2017 on behalf of Mr Gary Ekert and Mrs Kim Ekert of the proposed Bulahdelah Caravan Park This report wasprepared for the development of the Bulahdelah Caravan Park to ensure the flood risk for the visitors and staff is managed appropriately.

Table 3.8 Modelled peak flood levels (m AHD) for a range of design events.

event

event

Table 3.9 Modelled peak flood levels (m AHD) for combined catchment and ocean derived design events

The caravan park is located in the middle of the township of Bulahdelah, 90 m upstream of the Old Pacific Highway Bridge. The Caravan Park is located on the banks of the Myall River, opposite the confluence with the Crawford River.

The study tabulated the flood levels in the vicinity of the Bulahdelah Caravan Park site based on Flood Appraisal (1991) and Myall River FRMS (2002) - see Table 3.10. Table 3.10 Flood levels in the vicinity of the Bulahdelah Caravan Park.

Hazard and hydraulic category maps revealed that most of the site lies within the high hazard area highlighting that occupation of the site poses a significant flood risk.

The study analysed the Bulahdelah Flood Appraisal outcomes on the 5% and 1% AEP water level time series at the Bulahdelah Bridge just downstream of the Caravan Park The analysis determined the average rate of 0.5 m/hr of water level rise for the 5% and 1% AEP design events Due to the insufficient data documented in the previous studies on extreme events, the study adopted an extreme rate of rising of 2 m/hr which is four times greater than the rate of rising of the 5% and 1% AEP design events

The study concluded that:

• There is sufficient time and a safe route to evacuate both the Stage 1 and 2 development of the Bulahdelah Caravan Park based on active monitoring of river water levels

• Raising of the proposed internal road connecting Stage 1 and 2 sites, to a finished level of 4.0 m AHD

• Adoption of the route east of the site through the main access onto Stroud Street, south to Bulahdelah Way and east to Meade Street was recommended.

• An evacuation procedure was recommended to be implemented based on a range of factors discussed in detail in the report.

© Crown 2021 Classification: Release by consent

4 Data collection and review

4.1 Water level and rainfall data

MHL manages a number of water level and near-real-time rainfall recording stations within or in the vicinity of the study area. These gauges are tabulated in Table 4.1 and their locations are illustrated in Figure 4.1. This figure also displays the locations of a number of points at which rainfall data were extracted from the radar data from TheWeatherChaser.com for the various calibration/validation events. The nearest daily rainfall monitoring locations are the gauges maintained by the Bureau of Meteorology (BoM) tabulated in Table 4.1 Recorded water level data at Bulahdelah and Bombah Point stations are presented in Figure 4.2 and Figure 4.3 Rainfall data is presented in Figure 4.4

© Crown 2021

Classification: Release by consent

Table 4.1 Water level and rainfall stations n vicinity of the study area.

Figure 4 2 Water level data recorded at Bulahdelah Station 209460

Figure 4 3 Water level data recorded at Bombah Point Station 209475

Figure 4 4 Near-real-time rainfall data recorded at Bulahdelah Station 209460

4.2 Site inspection

A site inspection was carried out by the MHL project team on 1st of October 2020. The purpose of this inspection was to gain an overall appreciation of the relevant characteristics of the area and to identify areas that either contributed to flood risk or that were subject to the greatest flood risk. The site visit was undertaken with MidCoast Council Staff. A preliminary 1% AEP flooding event was run prior to the site inspection to understand the main flow paths within the study area. A number of pipes and culverts were then inspected along the main flow paths to determine if the size of the barrels as provided within the Council’s GIS layers was correct. Figure 4.5 presents the various locations that have been inspected during the site inspection. The majority of the culverts and pipes inspected appeared to be of a correct dimension but a number of discrepancies were observed as highlighted by the ten items numbered on Figure 4.5 and these discrepancies are listed below:

• No pipes were going down River Street (item 1), Myall Street (item 2) and Edgar Street (item 3) but instead there is a small dish drain.

• The pipe across Richmond Street near Edgar Street (item 4) appeared to be 450 mm instead of 375 mm.

• A small pipe across Richmond Street near Prince Street appeared to be 375 mm instead of 300 mm and another pipe across Prince Street near Richmond Street (item 5) was found to have two 900 mm cells instead of one.

• The culvert across Richmond Street halfwaybetween Prince Street and Alexandra Street (item 6) appeared to be 2 × 1250mm × 650 mm instead of 2 × 650mm × 375 mm.

• The culvert across Bulahdelah Way near Richmond Street (item 7) appeared to be a 3 × 1250 mm × 1250 mm instead of a single 375 mm pipe.

• The culvert across Bulahdelah Way halfway between Blanch and Stuart Streets (item 8) appeared to be a 2 × 1500 mm × 1250 mm instead of a single 375 mm pipe.

• The culvert across Bulahdelah Way near Rosewood Street (item 9) appeared to be 2400 mm × 900 mm instead of 2400 mm × 1200 mm.

Discussion with a few local community members highlighted some local flooding issues:

• Along Lee Street near the recently constructed roundabout with Stroud St.

• Along Jackson Street near Alexandra Street in the recent year possibly due to vegetation growing in the flow path.

Siteinspection

summary

Legend

Cadastre

Locationanddirectionof photographstakenduring siteinspection

MainFlowPaths

Pipes/CulvertsfromCouncil

ObservedDiscrepancy

ReportMHL2789

ReviewofBulahdelah

FloodplainRisk

ManagementStudy&Plan

4.3 Topographic data

Light Detection and Ranging (LiDAR) survey of the study area and its immediate surroundings was obtained for the study from ELVIS (https://elevation.fsdf.org.au/). 1 m resolution LiDAR data from the ‘Bulahdelah’ dataset (NSW Spatial Services August 2012, June 2013, January 2016 and January 2018) was used in the present study. The accuracy of the ground information obtained from LiDAR survey can be adversely affected by the nature and density of vegetation, the presence of varying terrain, the vicinity of buildings and/or the presence of water. The horizontal accuracy of the data is 0.8 m at 95% confidence interval, while the vertical accuracy is 0.3 m at 95% confidence interval The terrain topography is illustrated in Figure 4.6

Nearly several hundreds of survey marks were available from the Survey Control Information Management System (SCIMS) database developed by the NSW Government’s Spatial Services. The elevations of the survey marks extracted from the SCIMS online database were used to check the accuracy of the LiDAR data. The numerous survey marks were sorted to remove:

• Survey marks that were either damaged or not found

• Survey marks that had no elevation or only had round number value as they appeared to have been rounded to the nearest ten (e.g. 10 m, 30 m, etc.)

• One outlier with over 30 m level difference

Details of the survey marks including name, coordinates, elevation from SCIMS database, elevation measured from LiDAR data and difference in levels are provided in Appendix A

The survey marks locations and difference between SCIMS and LiDAR elevation are also illustrated in Figure 4 6 It can be noted that the survey marks with the largest differences are typically outside of the Bulahdelah township and located along steep surfaces. The LiDAR appears therefore adequate for the study.

4.4 Aerial photography

The most recent available aerial imagery was obtained from Google Earth (www.googleearth.com), captured in 2020 to observe current features within the study area. Some high resolution aerial imagery from 24 June 2020 was also available from Nearmap for the township of Bulahdelah.

4.5 Crawford River weir

A small weir is located on the Crawford River approximately 75 m upstream from the junction with the Myall River. The weir is a sheetpile structure reinforced with rock-fill and steel bracing and was constructed in 1964. The structure includes a fishway and is part of Bulahdelah water supply system. The crest of this weir is at approximately 0.75 m AHD and is visible in the LiDAR data. Photographs of the weir and fishway are provided in Figure 4.7.

4.6 Bridges

Council’s bridges drawings and GIS layer provided as a single point along Markwell Road, Markwell Back Road, Wootton Way and Old Pacific Highway were reviewed for completeness and quality for flood modelling purposes. The provided data includes:

• Nan Syron Bridge drawing including deck and underside levels, length, width, railing height, location and width of piers.

Figure 4 7 Crawford River weir and fishway (NSW Fisheries Office of Conservation, 2003)

• Battles Bridge drawing including deck and underside levels, length, width, railing height, location and width of piers.

• Frys Creek Bridge provided as a single point including approximate deck underside level, length, width.

• Markwell Back Road Bridge provided as a single point including approximate deck underside level, length, width.

• Wootton Way Bridge provided as a single point including approximate deck underside level, length, width.

MHL assumed the approximate deck underside level, length and width for the three doubleroad bridges along Pacific Highway based on the available topographic data and aerial photography.

4.7 Pits and pipe network

Council’s pit and pipe drawings were reviewed for completenessand quality for flood modelling purposes. The review identified some missing data required for modelling the storm-water network as follows:

• Pipe data including approximate locations and pipe sizing were available for the majority of the catchment.

• Minimum pipe size of 0.15 m was included in the hydraulic model located at the intersection of Crawford Street and Blanch Street, as well as at the intersection of Alexandra Street and Jackson Street.

• Information for several pipes was provided as a single point along Bombah Point Road, Markwell Rd, Markwell Back Rd, Suncrest Cl and Frys Creek Rd. Pipe alignment was therefore assumed and based on the available local topographic data.

• Pipesand culverts acrossPacific Highwayand Bulahdelah Way outside of the township were added to the model based on available Work-As-Executed drawings of the Bulahdelah Bypass developed by SMEC in November 2013 and from the Bulahdelah Upgrading the Pacific Highway Environmental Impact Statement Technical Paper 8 Water (RTA, 2004). A number of culverts were described as “Constructed by others” but the original dimensions from the drawing were assumed in the absence of further information. Only culverts and pipes crossing the highways were added to the model as they would have the largest impact on the flood modelling.

• Locations of pits were mostly available; however, invert level and pit sizing were not always present in the data provided.

MHL assumed a minimum cover depth of 100 mm to 700 mm for all provided pits using standard pipe grading as a guide to ensure hydraulic continuity. The inlets and outlets of the drainage system were checked to ensure that they are lining up with the available elevation data and aerial imagery (e.g. ensuring that headwalls are not located just outside of a dish drain).

All kerb-type pits were assumed to be 1800 mm wide and 200 mm high. A number of discrepancies in pipe sizes were highlighted during the site inspection as described in Section 4.2 The listed discrepancies were corrected prior to including the information into the flood model.

4.8 Floor level and post-event survey

MHL project team visited the Bulahdelah township between 30 March and 1 April 2021. This site inspection occurred after the 18-24 March large rainfall event that generated the largest water level recorded by the Bulahdelah gauge since its installation in late 1985. This site inspection had two main purposes:

• Measuring potential flood marks observed around the township

• Complete a visible survey of the floor level for the entire township of Bulahdelah

4.8.1 Post-event survey

A number of flood marks were measured during this site inspection including seven locations within Bulahdelah Township and one location upstream of Markwell Bridge. These flood marks were captured using a real-time kinematic GPS device (RTK) and are further discussed in the model calibration and validation section (Section 8.4).

4.8.2 Floor level survey

A total of 564 properties were surveyed for floor levels. The survey consisted of collecting the following information:

• Number of steps between the ground level and the floor level (assuming a 0.15 m step)

• Property type (e.g. residential, commercial, industrial, etc.)

• Construction type (e.g. slab-on-ground/low-set, high-set)

• Wall type (e.g. fibro, brick, timber, etc.)

• Number of storeys

• Photograph of the property

Results of the floor level survey are presented in Appendix B

4.8.3 Additional ground survey

During the same site visit, the project teams contacted the owner of the River Myall Holiday Resort and obtained permission to take a few survey points around the property to improve representation of the local topography into the hydraulic model since the existing LiDAR information was captured prior to the development of the property and changes in the local ground surface. These additional survey points were included into the model prior to running the design events.

5 Community consultation

5.1 Consultation process

Consultation provides an opportunity for various stakeholders, including the community, to collaborate in developing the Bulahdelah FRMS&P. Engaging the community throughout the process provides opportunities to both garner useful feedback and ideas regarding potential floodplain management measures, and increase community acceptance of the floodplain risk management plan.

The consultation program for the FRMS&P has included the following activities:

• Inception and progress meetings between the consultant and Council

• Meetings of the Bulahdelah Floodplain Risk Management Working Group

• Consultation with agencies and stakeholders

• Website development

• Letter and questionnaire for property owners

These activities are described at greater length below.

5.2 Working group

The Floodplain Management Advisory Committee (the Committee) was formed by Council to provide a forum that brings together the diverse expertise and community knowledge that is needed to address technical, social, economic and ecological issues concerning floodplain risk management in the study area. The committee fulfills the functions of a Floodplain Risk Management Committee as described in the Floodplain Development Manual (NSW Government, 2005).

The committee comprises representatives from:

• The community and farming industry

• Appropriate officers from Council and MidCoast Council Floodplain Management Advisory Committee, including planners and engineers

• NSW Department of Planning Industry and Environment, Biodiversity, Conservation and Science division (BCS)

• The NSWState Emergency Servicesand/or NSWPolice Force (acting in lieu of the NSW State Emergency Services due to the lack of Bulahdelah-based SES representative). It is noted that the nearest SES unit is located in Stroud. The NSW Police Force may receive assistance from the Rural Fire Service if required and are the main point of contact as they are located in town.

5.3 Website

A website was developed to provide information about the study, general flooding information, and a link to the online questionnaire Some snapshots of the project website are presented in

5.4 Letter and questionnaires

5.4.1 General approach

In November 2020, over 800 letters were distributed to all properties in Bulahdelah. The community letter alerted the residents and businesses of the online survey that was available to complete. A copy of the letter is included in Appendix C of this report.

An online survey was made available until end of January seeking community input about historic flooding and ideas about floodplain management options in the study area. The survey is included in Appendix C . A hardcopy of the survey was mailed to Bulahdelah’s residents at

Figure 5.1 Samples of the project website

the same time as the community letter.

The online survey has been using the SurveyMonkey platform.

5.4.2 Questionnaire results

A total of 59 responses were received. This represents a response rate in the order of 7%. A total of 17 responses were provided using the online questionnaire and 42 hard-copy responses were submitted. It is noted that the questionnaire was sent to all Bulahdelah residents which includes people that are in areas unlikely to be flood affected and who may be less likely to provide a response. Results of the survey are provided in Appendix C

The following key observation were made based on the community survey:

• The vast majority of respondents have a residential property (over 79.7%) 8.5% of the properties are farming/rural, 6.8% are vacant land and 3.4% are commercial.

• Over 84% of properties are owner occupied.

• Many residents have been living there for a long time (15.3% lived between 10-20 years, about 44.1% have lived there for over 20 years and 50.3% lived there for less than 5 years)

• About 50.8% of the respondent mentioned that their property was affected by flooding. It is important to note the personal interpretation of flood affectation (i.e. flooding of property vs. flooding of building).

• 25 respondents provided one or multiple example of flood events. The most commonly impacted part of the property included the ground (~41%), the garage/shed (~22%) and the access road (~19%). About 12% of the building were affected.

• Observed flood depths were typically described as over 0.3 m (~68% over 0.3 m, ~47% over 0.5 m, ~24% over 1 m and a couple of respondents mentioned depths greater than 3 m).

• Flooding durations range between less than 1 hour to several weeks with the majority (~76%) describing flooding as lasting for a number of hours up to 24 hours.

• The flood water was described primarily as having a running pace (~46%), followed by a walking pace (~34%) and then a stationary pace (~20%).

• The main sources of flooding were described as the water flowing from local drainage channel (~33%), floodwater rising from stream/river (~19%), water flowing down the roads (~19%) followed by overflow from neighbouring properties (~15%) and ponding of water on road (~11%).

• The expected response from the respondents during a future flood event would be to stay at the property (~64%), protect/raise/secure valuable and goods at risk (~24%), follow their flood management plan (~17%), evacuate early to official evacuation point (~12%) or move upstairs (~7%). About 12% of respondents were unsure of how they would react and ~7% did not specify any answer.

• The decision to evacuate or remain at the property will be made depending on if the house cannot be flooded/respondent can cope with isolation (~36%), family safety (~29%), availability of the access road (~25%), required access to medical facilities

(~20%), flood warnings (~17%), care for pets/animals (~14%) and discomfort/cost of being isolated by floodwaters (~7%).

• A number of residents provided photographs as well as other flood information. Example of photographs are provided in Figure 5.2 and Figure 5.3. Some photographs received from local residents were used for validation of the model (Appendix E).

Figure 5 2 Flooding along Lee Street (Photograph courtesy of Kevin Smith)

The main concerns and suggestions from the respondents regarding flooding included:

• Lack of drainage system maintenance and cleaning in the town (28 respondents)

• Requirement for upgraded drainage system (18 respondents)

• Channels and drainage system need to be cleared from vegetation (7 respondents)

• Impact from past development such as the Bulahdelah Bypass (3 respondents), the new

Figure 5.3 Flooding in Alexandra Street in 2020 (Photo Courtesy of Elfrieda McDean)

Figure 5.4 Full drainage channel looking downstream and upstream from Stroud Street in July 2020 (Photo Courtesy of John and Heather Lee)

caravan park (2 respondents) and local developments (2 respondents). NSW SES also noted that they attended local flash flooding in houses that had not previously been affected potentially due to some changes in local run off and drains flooding following the construction of the bypass.

• The need for kerb and gutter (2 respondents)

• Flooding from neighbouring properties (2 respondents)

• Weed control upgrade (2 respondents). A number of respondents suggested flood mitigations such as:

• Construction of a levee or filling land

• Flood zoning classification update

• Road surface upgrades

• Upgrade of the Richmond Street drain between Stroud Street and the river

• Deepening of drainage creeks

• Development control in flood affected areas such as Jackson Street

• Flood water diversion alongside Ann Street and Church Street directly to the river

• Piping of open drains

• Construction of swale or re-vegetation to slow down floodwater

• Excavation of silted Frys Creek near junction with Myall River

6 Hydrological analysis

Hydrologic modelling consists of determining the volume of water and the flows generated in a catchment based on various parameters including rainfall, catchment area, percentage of the ground that is pervious (such as grass or bare earth for example) or impervious (such as concrete or roads) and typical lag coefficient (which defines the time the flood water takes to travel through the catchment).

6.1 Model selection

The hydrological model selected for this study is WBNM (version 2017). This version of the model has been developed to include the 2016 Intensity-Frequency-Duration (IFD) diagrams that are at the basis of the ARR 2019 guideline requirements.

6.2 Model setup

6.2.1

Catchment delineation

This study utilised a WBNM hydrologic model that covers the catchment of approximately 436.8 km2 , shown on Figure 6.1. The catchment extends down to Bombah Broadwater. The catchment was divided into 213 sub-catchments, delineated based on an 8 m DEM developed from the available LiDAR information. The sub-catchments were delineated using CatchmentSIM version 3.6. This software was specifically developed to identify how subcatchments are connected and determine the surface characteristics of each sub-catchment such as area and percentage impervious.

6.2.2

Model parameters

Parameters required by the WBNM model include sub-catchment area and linkage, pervious and impervious percentage, runoff lag factor, stream routing lag factor, rainfall input, initial losses and continuing losses.

6.2.2.1 Imperviousareas

Impervious areas were derived by adopting impervious percentages for various land uses defined by Councils GIS layer for LEP zoning. A land use map is presented on Figure 2 10 Based on land use areas, a weighted average was calculated for each sub-catchment. The urbanized/residential areas were assumed to be 60% impervious, roadway corridors 100% and water bodies/basins 100% impervious while the rest was assumed as pervious surface. Table 6.1 summarises the percentage imperviousness used for each sub-catchment as per Figure 6.1

© Crown 2021

Table 6.1 Adopted impervious area percentage per sub-catchment

MHL2789 – 45

Classification: Release by consent

6.2.2.2 Losses

WBNM hydrologicmodelusesinitialand continuing rainfall lossesobtained from the ARR 2019 Data Hub (http://data.arr-software.org/).

Sensitivity of the flow to continuing loss for the total catchment upstream of Bulahdelah Bridge during the April 2015 event is illustrated in Figure 6.2. It can be noted that continuing losses require to be high to generate changes in the response in flows.

Based on the recent guideline developed by NSW DPIE (Office of Environment and Heritage, 2019), in the absence of calibrated losses (i.e. calibrated to flows at a stream gauge) in the catchment or nearby, the continuing loss value provided by the ARR 2019 Data Hub is to be multiplied by a factor of 0.4. In Bulahdelah catchment, the continuing rainfall loss is therefore taken as 0.4 × 3.1 mm/hr = 1.24 mm/hr. This value was compared to the continuing loss of 0.5 mm/hr (adopted in the Lower Myall River and Myall Lake Flood Study) as well as larger continuing loss values. The 0.5 mm/hr value provided the best match in the calibration and was therefore adopted for both calibration and design runs Moreover, a continuing rainfall loss of 0 mm/hr was adopted for impervious areas for both calibration and design events.

Sensitivity of the flow to initial loss for the total catchment upstream of Bulahdelah Bridge during the April 2015 event is illustrated in Figure 6.3. Larger initial losses would reduce peak flow and slow down the catchment flow response.

The initial loss for pervious area for calibration varies due to different antecedent conditions. The probability neutral burst initial loss values for pervious areas has been adopted for the design events (see Appendix D ). Moreover, an initial loss of 1 mm was adopted for

Figure 6 2 Sensitivity of flows to continuing loss for catchment upstream of Bulahdelah Bridge

impervious areas for both calibration and design events.

An initial loss of 0 mm and a continuing loss of 1 mm/hr were adopted for pervious areas for the PMF event as recommended in ARR 2019

6.2.2.3 Lag

WBNM recommends lag parameter values ranging between 1.3 and 1.8 with an average value of 1.6. It is also the recommended value for use on ungauged catchments for NSW (Boyd and Bodhinayake 2006). A stream lag routing Type R with a value of 1 was adopted. This is the recommended natural channel routing value.

Responses of the catchment upstream of Bulahdelah Bridge to changes in lag parameter C and routing parameter R during the April 2015 event are compared in Figure 6 4 and Figure 6 5 respectively. It is noted that both parameters have similar flow response patterns and the lag parameter C was therefore used as main calibration parameter.

It was found that a larger C parameter provided a slower response in flow with a less peaky rising limb and a flatter tailing limb of the hydrograph which is consistent with the water level response as measured at Bulahdelah.

A value of 2.5 was found to provide the best results. While this value is higher than the typical recommended values, most of the catchment is densely vegetated and a considerable portion of the catchment is very flat The riverbed is also very deep and low-lying which would significantly reduce the flow conveyance and therefore an increased lag would occur

Figure 6.3 Sensitivity of flows to initial loss for catchment upstream of Bulahdelah Bridge

Figure 6 4 Sensitivity of flows to Lag parameter for catchment upstream of Bulahdelah Bridge

Figure 6.5 Sensitivity of flows to Routing parameter for catchment upstream of Bulahdelah Bridge

6.2.3 ARR 2019 IFD data

The Australian Rainfall and Runoff (ARR) guidelines were updated in 2016, and revised in 2019, due to the availability of numerous technological developments, a significantly larger dataset since the previous edition (1987) and the development of updated methodologies. A key input to the process is information derived from rainfall gauges, and the dataset now includes a larger number of rainfall gauges, which continuously recorded rainfall (pluviometers), and a long record of storms, including additional rainfall data recorded between 1983 and 2012.

ARR 2019 recommended implementing the following features:

• Intensity, Frequency and Duration (IFD) rainfall data, pre-burst, and initial and continuing loss values across Australia;

• Critical duration derived from an ensemble assessment of 10 temporal patterns for each storm duration. The temporal pattern producing the mean level within each duration is selected. The critical duration is the duration for which the selected temporal pattern produces the maximum flood level;

• The inclusion of Areal Reduction Factors (ARFs) based on Australian data for short duration (12 hours and less), long duration (larger than 24 hours), and durations between 12 and 24 hours.

In the present study, the methodologies provided in ARR 2019 were implemented for design flood modelling, as these methodologies represent the best practice and would increase the longevity of the outputs of the Study.

6.2.4 ARR 2019 spatially varied IFDs

Design rainfalls (ARR 2019 IFDs) were obtained from the Bureau of Meteorology (BoM) for specific AEP and duration combinations across the catchment. The Data Hub metadata is presented in Appendix D

6.2.5 ARR 2019 temporal pattern

Temporal patternsdescribe how rain fallsovertime and form a component of storm hydrograph estimation. ARR 2019 has adopted a regionally specific ensemble of ten different temporal patterns for a particular design rainfall event. Given the rainfall-runoff response is catchment specific, using an ensemble of temporal patterns attempts to produce the median catchment response.

The ARR 2019 method provides patterns for 12 climatic regions across Australia, with Bulahdelah catchment falling within the Southern Semi-arid region. ARR 2019 provides patterns for each duration which are sub-divided into temporal pattern bins based on the frequency of the events. Figure 6 6 shows the categories of bins (very frequent, frequent, rare, very rare and extreme) and corresponding AEP groups. At the time of the model update, the “very rare” bin had been unavailable and was not used in this flood study; instead, temporal patterns from the “rare” bin were applied for the 1 in 200 and 1 in 500 AEP events. There are ten temporal patterns for each AEP/duration in ARR 2019 that have been utilised in this study for the 20% event to 1 in 500 AEP events.

Temporal patterns for this study were obtained from the ARR 2019 Data Hub (http://data.arrsoftware.org/). A summary of the data hub information at the catchment centroid is presented in Appendix D . The method employed to estimate the PMP utilises a single temporal pattern (Bureau of Meteorology, 2003), as is consistent with ARR 2019.

6.2.6 ARR 2019 areal temporal patterns

ARR 2019 recommends that Areal Temporal Patterns (ATP) should be considered in catchments larger than 75 km2 to account for the spatial smoothing of rainfall that occurs over larger catchments. Bulahdelah catchment covers an area of over 430 km2 which is above the threshold of considering ATPs. ATPs are applicable for storm durations including and greater than 12 hours. Hence, ATPs were applied when assessing longer durations above or equal to this threshold in the hydrologic model.

6.2.7 Areal reduction factors

ARFs are defined as an estimate of how the intensity of a design rainfall event varies over a catchment The main assumption in this estimate is that large catchments will not have a uniform depth of rainfall over the entire catchment.The ARFs were derived from the ARR 2019

Figure 6.6 Temporal pattern bins

Data Hub and applied in the WBNM models for all of the investigated design storm events. The ARF varies with AEP and duration and the resulting set of ARFs for the design storms are provided in Appendix D The ARF for the PMF has been set as 1 as per ARR 2019 recommendations.

6.3 Comparison with “At Site” and ARR 1987 data

The 2016 IFD curves from the Bureau of Meteorology were compared to the IFD created using the “At Site” rainfall gauges available in the vicinity of the study and the results are presented on Figure 6.7. The Bulahdelah Station (Station ID: 209460) had the highest sampling frequency and was used to determine the “At Site” IFD for events up to the 5% AEP. The relatively short period of record did not allow estimates of rarer events. It is noted that the 2016 and the “At Site” IFDs are typically consistent with the largest difference observed for durations of less than 1 hour.

The 2016 IFD were also compared to the 1987 IFD and the results are presented on Figure 6.8. It is observed that the 2016 IFD rainfall intensities are consistent with the 1987 IFD for frequent events and for long durations. The 2016 IFD rainfall intensities are higher than the 1987 IFD intensities for 10% AEP and rarer events particularly for shorter durations.

The 2016 IFD were adopted for this study as per ARR 2019 requirements.

Figure 6.7 Comparison of “At Site” IFD and 2016 IFD from BoM

6.4 Design events

The design events modelled in this study include:

• Frequent events – 20% AEP

• Rare events – 5% AEP and 1% AEP

• Very rare events – 1 in 200 AEP and 1 in 500 AEP

• Extreme event – Probable Maximum Flood (PMF)

The terminology of these events is defined as per the ARR 2019 guidelines presented in Figure 6.6. All events (except the PMF) use spatial and temporal patterns provided by the ARR 2019 Data Hub. The PMF uses a combination of other temporal and areal patterns as described in the following section.

6.5 Probable Maximum Flood event

The Probable Maximum Flood (PMF) is the largest flood event resulting from the Probable Maximum Precipitation (PMP). The PMP rainfall depth has been estimated using the ARR 2019 guidelines. According to the PMP method zones diagram (Bureau of Meteorology, 2003) Bulahdelah catchment fall within the GSAM-GTSMR Coastal Transition Zone. Therefore, durations of up to 6-hours have been considered for the PMP in accordance with the Generalised Short Duration Method (GSDM) derived by the Bureau of Meteorology (BoM) (Bureau of Meteorology, 2003) and durations of 24 hours or longer have been estimated using both the Generalised Southeast Australia Method (GSAM) (Bureau of Meteorology, 2006) and the Generalised Tropical Storm Method (GTSMR) (Bureau of Meteorology, 2005) Intermediary durations (i.e. 9 hr, 12 hr and 18 hr) have been estimated using all three methods and the maximum has been used for the PMF. A summary of the GSDM, GSAM and GTSMR results were provided in Table 6.2, Table 6.3 and Table 6.4, respectively

Figure 6.8 Comparison of 1987 IFD and 2016 IFD from BoM

Table 6.2 GSDM summary for Bulahdelah Catchment

Table 6.3 GSAM summary for Bulahdelah Catchment

Table 6.4 GTSMR summary for Bulahdelah Catchment

The temporal patterns used to derive the probable maximum flood (PMF) should be selected from an ensemble of patterns appropriate for use with the Generalised Probable Maximum Precipitation (PMP).

At present, the best source of ensemble temporal patterns for use with short duration PMF events are those derived by Jordan et al (2005) for durations up to 6 hours. The procedure suggested to derive the design temporal distribution of GSAM and GTSMR patterns for duration of 24 hours or longer were described in the revised edition of Australian Rainfall and Runoff, Book IV, ARR (Nathan and Weinmann, 1999) using the Average Variability Method (AVM) of Pilgrim et al., (1969) (Bureau of Meteorology, 2006) and (Bureau of Meteorology, 2005). The GSDM, GSAM and GTSMR patterns were used for intermediary durations (i.e. 9 hr, 12 hr and 18 hr). The Jordan et al. (2005) patterns were derived specifically from storms associated with thunderstorm or deeply convective events while the GSDM, GSAM and GTSMR patterns are defined in the associated guidelines The ellipse approach from the GSDM was applied to define the areal pattern for the shorter duration events.

These patterns were therefore adopted in this study and applied to the calculated PMP rainfall depth. The critical pattern was determined as per the typical ARR 2019 guidelines applied to the other design events.

6.6 Model results and critical durations

For each design AEP event, 23 different durations were modelled ranging from 10 minutes to 168 hours, except for the PMF which had twenty durations ranging from 15 minutes to 120 hours. Within each duration, 10 specific rainfall events were modelled (as recommended in ARR 2019) which varied the rainfall temporal pattern, though not the magnitude, over that period. This led to 230 individually modelled rainfall events per design event which were then analysed to pick the one event to use as design rainfall.

Critical durations were selected based on the methodology described in ARR 2019. This methodology consists of selecting, for each duration, the rainfall temporal pattern that is the closest to the average flow obtained from the 10 specific patterns provided in the ARR 2019 database. This provides an automated approach that can then be adjusted for consistency in durations between the various events.

Figure 6 9 and Figure 6 10 present box-plots for the 1% AEP for both the river catchment (sub-catchment 1.31) and for the township based on the local sub-catchment (sub-catchment

76.01) Critical durations and temporal pattern (from ARR 2019 Data Hub) based on the hydrologic model for each event are presented in Table 6.5. The design storm for each AEP is the rainfall event which results in the median flow for the critical duration (Figure 6.11). The selection of the critical duration for the overall Myall and Crawford Rivers catchments was based on the peak flow at Bulahdelah Bridge and an example is presented in Figure 6.12 for the 1%AEP event.

Figure 6.9 Critical temporal patterns (TP) distribution for the 1% AEP event at the township catchment (sub-catchment 76.01)

Figure 6.10 Critical TPs distribution for the 1% AEP event at the river catchment (sub-catchment 1.31)

Figure 6.11 Design storm selection from 10 TPs for 1% AEP, 18-hour storm (sub-catchment 1.31)

Figure 6.12 TPs distribution for all durations for the 1% AEP event at the river catchment (sub-catchment 1.31)

Table 6.5 Critical durations for each design event

* Figure 6.10 shows that pattern 112 is the critical duration. This pattern was compared to pattern 114 and the difference in flow is negligible and therefore, pattern 114 was used for all events to improve consistency between events and minimise the risk of having local areas being subject to lower flooding during rarer event due to change in temporal pattern.

** Temporal pattern ID from Jordan et al. (2005) rather than the data hub as recommended for PMF analysis in ARR 2019

6.7 Discussion on Crawford and Myall Rivers flood coincidence

Both the Crawford River and the Myall River catchments are compact and very close. They are therefore expected to typically be subject to similar rainfall patterns and to flood simultaneously. However, the Crawford River catchment was found to be approximately half of the Myall River catchment. This surface area difference would impact on the critical duration of each catchment and the Crawford River would most likely have a shorter critical duration than the Myall River. Application of the typical ARR 2019 approach of analysing all temporal patternsfor all durations would highlight which of the critical durations would lead to the highest flow at Bulahdelah. The temporal pattern and critical duration analysis are described in Section 6.6 The critical durations for the river catchment presented in Table 6.5 include the combined flows of the Myall River and the Crawford River.

The critical duration for the Myall River and Crawford River were then compared separately based on sub-catchments 1.29 for the Myall Riverand 56.11 for the Crawford River and results are presented in Table 6 6 This table compares both rivers’ critical durations and the peak flows from the Crawford River for the Myall River’s critical durations. It was found that both rivers have very similar critical durations. The critical duration of the Myall River only is the same as the combined rivers’ critical duration of 18 hours for most events and 6 hours for the PMF. The critical duration of the Crawford River was found to be between 12 and 24 hours with the 18-hour duration having very similar peak flows. It can also be noted that the flow from the Crawford River catchment is approximately half of the flow from the Myall River only catchment (hence a third of the combined flow at Bulahdelah). Both rivers are therefore likely to peak at the same time except for a very concentrated and localised storm cell.

Table 6.6 Critical durations comparison between the Myall River and the Crawford River

7 Hydraulic analysis

Hydraulic modelling consists of understanding the physical properties of the flood water such as depth and velocity. This can be completed in various ways including:

- One-dimensional (1D) modelling, which consists of representing a creek or river with flood information provided at regular interval cross-sections along a stream as well as pipe systems and drainage networks

- Two-dimensional (2D) modelling, which consists of representing a floodplain as a grid with flood information provided at each cell of the grid allowing the model to define flowpaths

- Two-dimensional modelling, which can be completed as a combination of 1D and 2D models.

7.1 Model selection

TUFLOW HPC has been used for hydraulic modelling in this study to simulate flood behaviour across the study area. TUFLOW is a robust and widely accepted unsteady-state flood simulation software with combined 1Dand 2D capabilities. The use of a TUFLOWmodel allows integrated investigation of local overland flow flooding, mainstream creek flooding, foreshore flooding and tidal influences, and the inclusion of storm-water drainage infrastructure.

The GIS data layers and control files used to drive the model can be easily modified for use in the optionsassessment, including modelling the impact of mitigation measures, or assessment of development applications. MHL flood modelling processes follow guidance provided in ARR 2019.

The dynamically linked 1D/2D model requires a number of GIS data layers to represent the study area. These include:

• 1D Domain

- Pits and headwalls GIS layer;

- Pipe network GIS layer;

- Culverts GIS layer;

• 2D Domain

- 2D grid / digital elevation model (DEM);

- Topographic modifications and break lines (e.g. to incorporate channel crosssections);

- Materials layer (specifies surface roughness and infiltration);

- Rainfall on the grid;

- Layered flow constrictions layer for 2D bridges; and

- Initial water level polygons

7.2 Model setup

The following considerations were required to set up the TUFLOW model.

7.2.1 Model extent and grid size

Testing of various model extents showed that the TUFLOW model was significantly impacted by the representation of the storage area within the floodplain. For this reason, the TUFLOW model was therefore set up to cover the majority of the floodplain area below approximately 10 m AHD. This covers an area of approximately 89 km2 between Bombah Broadwater and approximately 5 km upstream of Bulahdelah. Given the area of this model, the Quadtree capability of TUFLOW was adopted to represent the catchment. This capability allows the use of multiple cell sizes in the model to increase resolution in areas of interest by halving the cell size a number of times. The grid definition adopted for the TUFLOW model is presented in Figure 7.1

.

A grid cell size of 3.0 m by 3.0 m was found to be suitable to represent flooding within the township of Bulahdelah and was also applied to represent embankment structures such as elevated roads (blue extent in Figure 7.1) The Quadtree capability was then used to transition to a 6 m cell size in the floodplain where rural properties are located and along the main river (yellow extent in Figure 7.1), and to 12 m for the rest ofthe catchment that ismainly composed of large vegetated areas (red extent in Figure 7.1).

The sub-grid sampling (SGS) capability of the TUFLOW model was also set to 1 m (i.e. the resolution of the available DEM). The SGS allows the use of sub-grid-scale elevation data to enhance the hydraulic accuracyof the model (byproviding an improved representation of flows in and out of each cell and the definition of the volume within each cell) while keeping reasonable run times.

This variable size grid complemented by the activation of the SGS allows an appropriate representation of the features of the local urban catchment while keeping the run time reasonable. Initial timesteps of 1.5 second for the 2D model and 0.75 second for the 1D model have been adopted as these are the recommended values for a 3 m cell size (being the smallest cell size in the model) TUFLOW HPC uses an adaptive timestep approach to maintain stability and would vary this original value as required.

7.2.2 Modelling approach

MHL applied the following modelling approach to the development of a detailed, calibrated and reliable 2D/1D TUFLOW hydraulic model for the study area:

• Extent of the study area and 2D hydraulic model was determined based on the available elevation data

• Direct rainfall method was adopted over the 2D model extent and upstream catchments were represented as inflows calculated using WBNM

• Stormwater infrastructure: All pits, pipes, culverts and bridges were modelled as described in Section 7.2.6

• Blockages: The blockage applied to the pits and pipes system has been established by following the method described in the blockage assessment form provided in ARR 2019 and ARR Project 11: Blockage of Hydraulic Structures

• Hydraulic roughness: a materials layer was delineated from Council’s cadastre, zoning and aerial photography along with site observations. Initial material categories and associated depth-varying Manning’s roughness coefficients were established for the present study

• Fences: given significant uncertainty and variation in fence type, hydraulic behaviour and permanency throughout the study area, and a lack of verification of hydraulic modelling capabilities to represent them, MHL did not specifically model fences. Rather a ‘lumped’ hydraulic roughness approach was adopted

• Thorough model calibration, validation, verification by alternative methods and quality assurance checks.

7.2.3 Hydraulic roughness

Hydraulic roughness coefficients (Manning’s n) are used to represent the resistance to flow of different surface materials. Hydraulic roughness has a major influence on flow behaviour and is one of the primary parameters in hydraulic model calibration.

Spatial variation in hydraulic roughness is represented in TUFLOW by delineating the catchment into zones of similar hydraulic properties. The hydraulic roughness zones adopted in this study have been delineated based on consideration of Council LEP zoning, cadastral data and aerial photography. Factors affecting resistance to flow were of primary importance including surface material, vegetation type and density, and the presence and density of flow obstructions such as buildings, fences and gross pollutant traps (GPTs). Manning’s ‘n’ values assigned to each zone were determined based on aerial imagery, with reference to standard values recommended by Chow (1959). As resistance to flow due to surface and form roughness varies with depth (e.g. Chow 1959, Institution of Engineers Australia 1987), variable depth-dependent hydraulic roughness values have been adopted for this study to consider the typical sizes of vegetation/obstruction (i.e. typical grass or brush height) Once obstruction is underwater, roughness reduces. Figure 7.2 and Table 7.1 summarise the Manning’s n values used in the hydraulic model. The creek/river roughness used for the TUFLOW model was increased to consider the deep pools present along the river with large variations in invert level from -2 to -11 m AHD at regular intervals along the river channel.

N.B.: The Manning’s n value is changing with depth and for example, for Open Space, the Manning’s n value is 0.075 up to a depth of 0.1m and then transition to a smaller Manning’s n of 0.04 at a depth of 0.5 m or more.

Table 7 1 Adopted Manning’s n Hydraulic Roughness Coefficients

7.2.4 Boundary conditions

Key boundary conditions are illustrated in Figure 7.2 and further details are provided below.

Inflows

The simulated hydrographs produced by the WBNM model for the wider catchment upstream of the TUFLOW model were input at key locations at the upstream end of the TUFLOW model as shown in Figure 7.2

Moreover, within the TUFLOW domain, the direct rainfall approach was applied to the model using the design events identified and defined in the hydrological analysis (Section 6).

Downstream boundary condition

The water level as recorded at the Bombah Point water level gauge was applied as downstream boundary condition for the calibration and validation runs.

The Bombah Point gauge data was analysed to determine appropriate downstream water levels for the design events.

Table 7.2 summarises the peak water levels in Bombah Broadwater based on the annual maximum series (AMS) probability estimation of the water levels at the Bombah Point gauge and as extracted from the 2015 Lower Myall River and Myall Lakes Flood Study It is noted that the period of record is only 20 years at the Bombah Point gauge and the analysis can be influenced by large flood events. For example, the March 2021 level at Bombah Point was much larger than the other values recorded since the installation of the AMS analysis gauge back in July 2001 and this may skew the results of the AMS results over such a short record period. The AMS analysis was therefore completed with and without this event

Table 7.2 Peak design levels at Bombah Point

at Bombah Point gauge / in Bombah

by:

* Annual Maximum Series based on 20 years of data only

** Value was interpolated based on other values provided in flood study

Based on this comparison, the 2015 study values were adopted as tailwater for this study for the following reasons:

• The 2015 study values are based on modelled levels rather than raw data analysis and noting that higher levels within the lake would be subject to the topography and increasing volume of the lakes which the AMS does not accurately consider; and

• The AMS was only completed over a 20-year period of record allowing the data to potentially be skewed by large events with return period larger than 20 years that occurred over this period.

Based on the historical water level and rainfall data measured at Bulahdelah and Bombah Point, it was noted that the flooding of Bulahdelah and Bombah Point are not independent and are likely to coincide as the Myall River catchment contributes to a significant portion of the catchment leading to Bombah Broadwater. Therefore, coinciding AEPs have been assumed for Bombah Broadwater and the Myall River (i.e. a 1% AEP design flood event for Bulahdelah was associated with a 1% AEP water level at Bombah Broadwater).

However, further analysis was completed on the timing of the peak of flooding at the Bulahdelah and Bombah Point gauges. It was found that the peak level at Bombah Point typically occurs several hours after the peak at Bulahdelah. This was considered when developing the tailwater level behaviour for each flood event.

To estimate a tailwater behaviour for the model, the following process was followed:

• the volume of the lake at 0.1 m intervals was estimated up to 3 m AHD, as well as at 4 m and 5 m AHD, based on the topographic (LiDAR) data to create a simplified “bucket” model of the lake with an outflow from the lake that was estimated based on these volumes and the recession rate observed during the largest event on record at Bombah Point (i.e. March 2021).

• The inflows calculated in the WBNM model were then inputted in this model (assuming a typical initial water level of 0.5 m AHD for design events) to determine a general shape for the rising limb of Bombah Point based on the Bulahdelah inflows.

• This general shape of the rising limb was then scaled to peak at the appropriate level at Bombah Point foreach design event (e.g. peakof 2.38 m AHD for a 1%AEP flood event). This approach was applied to the March 2013 and March 2021 calibration/validation events and the results are provided in Figure 7.3 While the full timing is not an exact match, it appears to follow a relatively similar trend, and this was considered acceptable for the purpose of defining a tailwater level at the downstream end of the TUFLOW model. This is particularly true given the relatively small influence of the tailwater level on the peak water level at Bulahdelah as discussed in Section 9.1. Therefore, this approach was applied to define the tailwater for the longer critical duration associated with the flooding of the river. The peak Bombah Broadwater values were directly used as a constant value for the shorter critical duration associated with the flooding of the township.

Figure 7.3 Comparisons of recorded data with tailwater level estimated using WBNM for March 2013 and March 2021 events

7.2.5 Blockage

Bridges and culverts are structures that allow water to flow under roads, railways or other obstruction from one side to the other. These structures can be affected by various blockage mechanisms, resulting in increased flood levels, changes to stream flow patterns, changes to erosion and deposition patterns in channels, and physical damage to the structures. Blockage of these structures is discussed in ARR 2019.

This ARR 2019 blockage procedure presented in the Blockage Assessment Form was followed. Cross-drainage structures were identified from Council GIS and included the pedestrian footbridge across Frys Creek.

Each cross drainage was assigned a “High”, “Medium” or “Low” rating for the following ARR 2019 attributes:

• Debris availability – this rating was based on aerial imagery to assess the upstream catchment and the availability of debris

• Debris mobility – this rating was defined using contours based on steepness of the source area and proximity of source area to streams.

• Debris transportability – based on stream dimension in comparison to potential debris as well as stream shape.

• Debris length L10 : ARR 2019 defines this value as:

- the average length of the longest 10% of the debris reaching the site and should preferably be estimated from sampling of typical debris loads. However, if such data is not available, it should be determined from an inspection of debris on the floor of the source area, with due allowance for snagging and reduction in size during transportation to the structure.

- In an urban area the variety of available debris can be considerable with an equal variability in L10. In the absence of a record of past debris accumulated at the structure, an L10 of at least 1.5 m should be considered as many urban debris sources produce material of at least this length such as palings, stored timber, sulo bins and shopping trolleys.

- A value of 1.5 m has been adopted for L10 for all blockage structures in the model. Based on the above approach, the majority of cross-structures have opening lower than the selected L10 and have a design blockage varying between 25% and 100% depending on the AEP ofthe event. The pedestrian bridge west of Rose Street and the culvert under David Street have a larger opening and the blockage would vary between 10% and 20% depending on the AEP of the event.

Regarding all other pits and pipes that were not identified as cross drainage structure, a 50% design blockage was adopted. This value was developed based on the design blockages described in Table 7.1 of ARR Project 11 – Blockage of Hydraulic Structures Stage 2.

7.2.6 Structures

In accordance with the available structures data and topographic data the following structures were included in the hydraulic model:

• Nan Syron Bridge was modelled at 1D elements

• The edge of the Bulahdelah Bypass bridge was modelled as a form loss into the 2D model while the centre of the bridge above the 1D river was not included as the deck level was well above the PMF level.

• Battles Bridge, Frys Creek Bridge and three doubled bridges along Pacific Highway were modelled as 2D elements

• Markwell Back Road Bridge and Wootton Way Bridge were modelled as 1D elements.

• The Crawford River weir was modelled in 2D based on LiDAR levels.

• All pipes and culverts within the study area were modelled as 1D elements.

Figure 7.4 illustrates the pipe network, culverts and bridges included in the hydraulic model.

7.2.7 Creeks and river representation

Following the site inspection, the upstream creeks appeared to be relatively shallow. Some information on depthsand historical cross-sectionswere made available and used to represent the various creeks and rivers. The following assumptions were made in the model:

• Several historical cross-sections of the Myall River were provided as PDF drawings from the Myall River Flood Study in May 1990 and a few cross-sections were also available from the 1991 Bulahdelah Flood Appraisal report. The various cross-sections were used to represent most of the Myall River channel between Bombah Broadwater and Bulahdelah as a 1D model.

• The upstream sections of the various creeks and the upstream end of the Myall River were assumed to be approximately 0.5 m deep and the lowest point observed from the LiDAR data was lowered by 0.5 m assuming a V-shape of the channel from the bank to the invert level.

• No cross-section of Frys Creek was available. However, an invert level was available at regular intervals along the creek (Public Works, 1994), and this level was assumed along the creek, and a V-shape channel was assumed in the upstream creeks.

• The Crawford River channel was reported to be in the order of 3 m deep (NSW Fisheries Office of Conservation, 2003). Therefore, the invert level was set at -2.5 m AHD at the upstream end of the model deepening to -2.7 m AHD at the junction with the Myall River. Based on aerial photograph and LiDAR data, the deep centre of the channel was assumed to be approximately 12 m wide and then transition to the existing bank level where LiDAR becomes representative (i.e. where no water is present).

• Some minor channels leading to ponding areas along the river were also adjusted to remove anomalies in the LiDAR data where blockage of the channels was indicated.

• The DEM at the downstream end of the Myall River, where an anabranch is located, was adjusted to include some local storage assuming a transition from -5 to -6 m AHD depths along the Myall River to a depth of -2 m AHD within the anabranch.

• The ground elevation of the caravan park directly north of the Nan Syron Bridge was updated based on the survey data collected during the project team site visit discussed in Section 4.8.3

8 Model calibration and validation

8.1 Methodology

Model calibration is an essential step in the flood modelling process to confirm that the model can adequately simulate historical flood events. In order to carry out model calibration, it is necessary to have available suitable recorded data sets against which to evaluate model results. The selection of appropriate historical events for model calibration is, therefore, largely dependent on the availability of relevant flood data.

The most reliable recorded flood data available in the study area are the Bulahdelah and the Crawford River water level records. Although recorded depth and flow data are not available higher in the catchment, calibration against the Bulahdelah and Crawford River water level gauges would indicate that the overall model performance within the Myall River catchment is reliable, and provides confidence in the model parameters and data being adopted throughout the study area.

The Bulahdelah and Crawford River water level records, together with recorded rainfall data, therefore, act as the primary basis for model calibration, with anecdotal flood depth data collected through community consultation also utilised.

8.2 Event selection

Suitable historical calibration and validation events were determined through considering the following criteria:

• the availability of Bulahdelah and Crawford River water level and continuous rainfall data

• the historical significance of recorded rainfall;

• the influence of recorded rainfall on the Myall River water levels;

• the availability of flood depth data collected through community consultation; and

• the representation of the development in the catchment (i.e. events post-construction of the Bulahdelah Bypass were preferred).

A review of the available historical information highlighted the 1-4 March 2013 event as the most suitable event for model calibration because it was the second most significant rainfall event recorded since installation of the Bulahdelah water level gauge in 1985, it was postconstruction of the Bulahdelah Bypass, and the rainfall pattern appeared relatively consistent throughout the catchment

The 21-22 April 2015 event was selected for model validation as this was the most significant rainfall event with Bulahdelah water levels available at the commencement of the study (i.e. prior to the March 2021 event), that occurred after the construction of the Bulahdelah Bypass, and a number of photographs were provided by the community for this event.

Given that the 18-24 March 2021 flood event was the most recent event as it occurred during the development of the flood study and recorded the highest flood level captured by the Bulahdelah gauge since its installation in late 1985, this flood event was used as a secondary validation event.

8.3 Model calibration – 1 to 4 March 2013

8.3.1 1-4 March 2013 event

Rainfall in the study area on 1-4 March 2013 fell primarily over a duration of approximately 24 hours, with a relatively consistent rainfall intensity. This led to an elevated flood level above 3.5 m AHD in the Myall River and at Bulahdelah lasting for over a day. A time-series plot of water level and rainfall data recorded during the event is presented in Figure 8.1.

8.3.2 Rainfall data

Only the Bulahdelah continuous rainfall gauge was active within the study area on 1-4 March 2013. However, a number of neighbouring daily gauges were available around Bulahdelah including the Crawford River and The Cabbage Tree Mountain gauges. These daily gauges showed very similar total rainfall depth during the event and the Bulahdelah Rainfall pattern was therefore adopted for the entire catchment The Crawford River rain gauge was represented by scaling the total daily rainfall measured during this event at this gauge using the Bulahdelah gauge rainfall pattern

Rainfall data converted from radar data was obtained from TheWeatherChaser.com, however, a significant portion of the rainfall event was missing from the radar data and this information was therefore not usable for this event.

8.3.3 Intensity-Frequency-Duration analysis

In order to provide relative context to the intensity of the 1-4 March 2013 rainfall event, the maximum rainfall depth recorded over a given duration has been compared with design Intensity-Frequency-Duration (IFD) data for Bulahdelah, as shown in Figure 8.2.

Based on the Bulahdelah gauge, the 1-4 March 2013 event was found to represent an event equivalent to a 10-20% AEP design rainfall for durations of 30 to 36 hours

Figure 8.1 1-4 March 2013 Event Time-Series

8.2 IFD Comparison 1-4 March 2013

8.3.4 Rainfall loss parameters

The translation of rainfall into runoff is directly influenced by the antecedent soil moisture conditions throughout the catchment. Rainfall losses are applied in hydrologic modelling to represent the amount of rainfall that does not contribute to runoff, primarily as a result of infiltration processes. The initial loss-continuing loss approach is widely accepted and was adopted in this study.

It should be noted that the initial loss-continuing loss model has been developed for use in traditional hydrologic models, which do not consider losses associated with topographic features, for example depression storage. The direct rainfall approach used in this study applies rainfall directly to each model cell and generally results in initial losses associated with small ‘pits’ in the DEM. Research has shown that such losses can be of the same order as traditionally adopted initial loss values (Taaffe et al. 2011).

There have been regular small rainfall events prior to this event and initial losses of 0 mm were applied in the WBNM and TUFLOW model for this event. A continuing loss value of 0.5mm/hr was adopted for pervious areas as previously used in the Lower Myall River and Myall Lake Flood Study It is noted that a 2.5 mm/hr continuing loss was previously used in the past Bulahdelah flood study, however this value appeared to be underestimating the peak of the flood. No losses were applied to impervious or permanently wet areas.

8.3.5 Upstream boundary condition

The flows resulting from the WBNM for the wider catchment were applied at key locations at the upstream end of the TUFLOW model as described in Section 7.2.4.

8.3.6 Downstream boundary condition

The Bombah Point water level gauge recorded data was used as downstream boundary condition for this event.

8.3.7 Rain radar observations

Newcastle rainfall radarobservationsasobtained from TheWeatherChaser.com website which provides access to the Bureau of Meteorology rain radar archives are summarised in

Figure

Appendix E. This website allows the extraction of marker intensity timeseries at selected locations as illustrated in Figure E.1 and the progression of the rainfall radar at regular (1hour) intervals as shown in Figure E.4 The marker intensity timeseries present the rainfall intensity between two dates in a colour bar format. The markers were extracted at three locations: the Bulahdelah Bridge Gauge (BBG), in the Upper Myall River Catchment (UMR) and in the Upper Crawford Catchment (UCR) The three markers were aligned to compare the intensity over time at the three locations. It is noted that the Bulahdelah Bridge gauge is slightly more intense at the start of the storm and less intense at the end of the storm when compared to the Upper Crawford River catchment. The Upper Myall Catchment is not severely impacted by this storm.

8.3.8 Model calibration results

A comparison of recorded and simulated water levels at the Bulahdelah and Crawford River gauges for the 1-4 March 2013 event is shown in Figure 8.3. A strong correlation between simulated and recorded water levels has been achieved in terms of:

• Peak water levels at Bulahdelah gauge – peak flood levels show a fair agreement for this event. The modelled level at peak is within 0.08 m of the maximum recorded level.

• Peak water levels at Crawford River gauge – peak flood levels show a good agreement for this event. The modelled level at peak is within 0.03m of the maximum recorded level.

• Catchment runoff response – the relative timing of the recorded and simulated water levels also shows a strong agreement on the rising limb of the flood. The slight difference in the falling-limb may be related to the Bulahdelah temporal pattern being the only temporal pattern available and being applied for the entire catchment. This can be seen on the marker intensity timeseries in Figure E.1 in Appendix E The Bulahdelah Bridge marker shows a more intense rainfall at the start of the storm and less intense at the end of the storm in comparison to the Upper Crawford River marker. This explains the slightly peakier hydrograph and faster tailing limb, and why the modelled Crawford River gauge responds slightly faster at the start of the storm.

The results of the 1-4 March 2013 event provide confidence in the ability of the developed TUFLOW model to simulate actual catchment flood behaviour.

Figure 8 3 Bulahdelah and Crawford River Water Levels Calibration – 1-4 March 2013

8.4 Model validation – 21 to 22 April 2015

8.4.1 21-22 April 2015 event

The rainfall leading to flooding in Bulahdelah on 21-22 April 2015 fell primarily over a duration of 3 to 6 hours, leading to a relatively short spike in water level in the Myall River. A time-series plot of water level and rainfall data recorded during the event is presented in Figure 8.8.

8.4.2 Rainfall data

Only the Bulahdelah continuous rainfall gauge was active within the study area on 21-22 April 2015. However, a number of neighbouring daily gauges were available around Bulahdelah including the Crawford River, Wootton and the Cabbage Tree Mountain gauges. These three gauges were applied the same temporal pattern as Bulahdelah and used as input for the modelling to consider the spatial distribution of the rainfall event. Cumulative rainfalls recorded at these gauges for the event (based on the total daily rainfall and the Bulahdelah Rainfall gauge temporal pattern) are shown in Figure 8.9.

Figure 8.4 21-22 April 2015 Event Time-Series

Figure 8 5 Cumulative Rainfall during the April 2015 event

Differences are evident in the rainfall records at each site. While these differences are likely to be the result of real spatial rainfall variability, it is notable that the Crawford River gauge recorded higher totals during the validation rainfall event. The majority of recorded rainfall fell over a period of approximately 6 hours from 18:00 on 20 April 2015 to 0:00 on 21 April 2015.

Rainfall data were extracted from the radar data and obtained from TheWeatherChaser.com at 10 locations around the catchment illustrated in Figure 4 1. This data was compared to the available rainfall data captured at the various gauges around the catchment It was found that while the actual rainfall depth extracted from the radar wasmuch lower than the recorded value at the gauges within the catchment, the general temporal pattern was fairly consistent with the recorded data. Therefore, the data from the variouspoints wasused to provide a local temporal pattern, and the rainfall depth was scaled based on the Bulahdelah gauge. This rainfall data was then adopted for modelling of this validation event as it provided a better resolution of the temporal pattern over such a large catchment.

8.4.3 Intensity-Frequency-Duration analysis

In order to provide relative context to the intensity of the April 2015 rainfall event, the maximum rainfall depth recorded over a given duration has been compared with design IntensityFrequency-Duration (IFD) data for Bulahdelah, as shown in Figure 8.10

Figure 8 6 IFD Comparison 21-22 April 2015

Based on the Bulahdelah gauge, the 21-22 April 2015 event was found to represent an event equivalent to a 10 to 20% AEP design rainfall for durations of 180 to 360 minutes. It is noted however that the daily rainfall measured at Crawford River gauge was about 2.5 times the depth of rainfall as Bulahdelah recorded on the same day and the AEP of this event is, therefore, likely to be greater.

8.4.4 Rainfall loss parameters

As shown in Figure 8.9, between 100 mm and 260 mm of rain had fallen over the study area

in approximately 12 hours leading to the event. Considering the dry antecedent conditions and the heavily vegetated upstream catchment, a 20 mm initial loss was applied in the TUFLOW model for this event. A continuing loss value of 0.5 mm/hr was adopted for pervious areas as per the March 2013 flood event. No losses were applied to impervious or permanently wet areas (i.e. road, concrete channel, creek and estuary zones).

8.4.5 Upstream boundary condition

The flows resulting from the WBNM for the wider catchment were applied at key locations at the upstream end of the TUFLOW model as described in Section 7.2.4.

8.4.6 Downstream boundary condition

The Bombah Point water level gauge recorded data was used as downstream boundary condition for the validation event.

8.4.7 Rain radar observations

Newcastle rainfall radarobservationsasobtained from TheWeatherChaser.com website which provides access to the Bureau of Meteorology rain radar archives are summarised in Appendix E. This website allows the extraction of marker intensity timeseries at selected locations as illustrated in Figure E.2 and the progression of the rainfall radar at regular (30minute) intervals as shown in Figure E.5. The marker intensity timeseries present the rainfall intensity between two dates in a colour bar format. The markers were extracted at three locations: the Bulahdelah Bridge Gauge (BBG), in the Upper Myall River Catchment (UMR) and in the Upper Crawford Catchment (UCR). The three markers were aligned to compare the intensity over time at the three locations. It can be observed that the Bulahdelah Bridge gauge was subject to a front loaded storm while the Upper Crawford River catchment is subject to a back loaded storm.

8.4.8 Model validation results Comparison with Recorded Water Level

A comparison of recorded and simulated water levels at the Bulahdelah and Crawford River gauges for the 21-22 April 2015 event is shown in Figure 8.7. This figure also presents the tailwater boundary condition calculated at the downstream end of the model. A strong correlation between simulated and recorded water levels has been achieved in terms of:

• Peak water levels at Bulahdelah gauge – peak flood levels show a fair agreement for this event. The modelled level at peak is within 0.13 m of the maximum recorded level.

• Peak water levels at the Crawford River gauge – peak flood levels show a relatively poor agreement for this event. The modelled level at peak is about 0.33 m lower than the maximum recorded level. This difference in peak is due to the limited rainfall data available across such a large catchment aswellas limited bathymetric information along the Crawford River

• Catchment runoff response – the relative timing and behaviour of the recorded and simulated water levels as well as the overall volume also show a general agreement on both the rising limb and the falling limb of the flood for both the Bulahdelah and the Crawford River gauges

The results of the 21-22 April 2015 event therefore provide confidence in the ability of the developed TUFLOW model to simulate actual catchment flood behaviour at Bulahdelah.

Comparison with flood photographs of the event

A number of photographs of the April 2015 flood event were provided highlighting flooding in Bulahdelah and the results of the calibration were compared to these photographs. The comparison is provided in Table E.1 of Appendix E. The model appears to reasonably represent the flowpaths and inundation extents highlighted by the provided photographs.

Figure 8.7 Bulahdelah and Crawford River Water Levels Validation – 21-22 April 2015

8.5 Model validation – 18 to 24 March 2021

8.5.1 18-24 March 2021 event

The rainfall leading to flooding in Bulahdelah on 18-24 March 2021 fell primarily over a duration of 4 days, leading to the largest water level recorded by the Bulahdelah gauge since its installation on the Myall River in 1985. A time-series plot of water level and rainfall data recorded during the event is presented in Figure 8.8 It can be noted that the event includes two peaks: a first peak occurred in the morning of Saturday 20 March and a second peak of similar amplitude occurred in the middle of the night from Sunday to Monday 21-22 March.

Event Time-Series

N.B.:ItisnotedthatBombahPointgaugewasfloodedduringthiseventanddatabetween6pmon21/3/2021and8pm on24/03/2021wereestimatedbasedontheavailabledataandtherecordedpeaklevelof2.89mAHD

8.5.2 Rainfall data

Only the Bulahdelah and Markwell continuous rainfall gauges were active within the study area on 18-24 March 2021. However, a number of neighbouring daily gauges were available around Bulahdelah including the Crawford River and the Cabbage Tree Mountain gauges.

Rainfall data was extracted from the radar data and obtained from TheWeatherChaser.com at 10 locations around the catchment illustrated in Figure 4.1. This data was compared to the available rainfall data captured at the various gauges around the catchment. It was found that while the actual rainfall depth extracted from the radar wasmuch lower than the recorded value at the gauges within the catchment, the general temporal pattern was fairly consistent with the recorded data. Therefore, the data from the various points were used to provide a local temporalpattern and the rainfall depth wasscaledbased on the Bulahdelah gauge. This rainfall data was then adopted for modelling of this validation event as it provided a good resolution of the temporal pattern over such a large catchment.

Cumulative rainfall recorded at these gauges for the event (based on the total daily rainfall and scaling of the rainfall radar information) is shown in Figure 8.9

Figure 8.8 18-24 March 2021

Differences are evident in the rainfall records at each site. While these differences are likely to be the result of real spatial rainfall variability, it is notable that the Cabbage Tree Mountain gauge recorded higher totals during this rainfall event.

The majority of recorded rainfall fell between 18 March and 22 March 2021.

8.5.3

Intensity-Frequency-Duration analysis

In order to provide relative context to the intensity of the March 2021 rainfall event, the maximum rainfall depth recorded over a given duration has been compared with design Intensity-Frequency-Duration (IFD) data for Bulahdelah, as shown in Figure 8 10

Based on the Bulahdelah gauge, the 18-24 March 2021 event was found to represent an event larger than a 5% AEP design rainfall for durations of 4 to 7 days, and about 20% for the critical

Figure 8.9 Cumulative Rainfall during the March 2021 event

Figure 8.10 IFD Comparison 18-24 March 2021

duration of the catchment of 18 hours. It is noted however that the daily rainfall measured at the Cabbage Tree Mountain gauge was about 2.5 times the depth of rainfall as Bulahdelah recorded on the same day; the AEP of this event is,therefore, likely to be greater. Furthermore, the very long duration led to significant flooding of the Bombah Broadwater, which reached a level of 2.89 m AHD and contributed to some backwater effect at Bulahdelah.

8.5.4

Rainfall loss parameters

As shown in Figure 8.9, between 420 and 770 mm of rain had fallen over the study area in approximately 4 days leading to the event. Considering the wet antecedent conditions, a 0 mm initial loss was applied in the TUFLOW model for this event. A continuing loss value of 0.5 mm/hr was adopted for pervious areas. No losses were applied to impervious or permanently wet areas (i.e. road, concrete channel, creek and estuary zones).

8.5.5

Upstream boundary condition

The flows resulting from the WBNM for the wider catchment upstream of the TUFLOW model were applied at key locations at the upstream end of the TUFLOW model as described in Section 7.2.4

8.5.6 Downstream boundary condition

The Bombah Point water level gauge recorded data was used as downstream boundary condition for the validation event. It is noted that the Bombah Point gauge was flooded during this event and data between 6 pm on 21/3/2021 and 8 pm on 24/03/2021 were estimated based on the available data and the recorded peak level of 2.89 m AHD.

8.5.7

Rain radar observations

Newcastle rainfall radarobservationsasobtained from TheWeatherChaser.com website which provides access to the Bureau of Meteorology rain radar archives are summarised in Appendix E. This website allows the extraction of marker intensity timeseries at selected locations as illustrated in Figure E.3 and the progression of the rainfall radar at regular (2hour) interval as shown in Figure E.6. The marker intensity timeseries present the rainfall intensity between two dates in a colour bar format. The markers were extracted at three locations: Bulahdelah Bridge Gauge (BBG), in the Upper Myall River Catchment (UMR) and in the Upper Crawford Catchment (UCR). The three were aligned to compare the intensity over time at the three locations. It can be observed that some parts of the catchment are impacted at different times and the rainfall can sometimes fall on localised parts of the catchment (e.g. upstream only or downstream only)

8.5.8 Model validation results Comparison with Recorded Water Level

A comparison of recorded and simulated water levels at the Bulahdelah and Crawford River gauges for the 18-24 March 2021 event is shown in Figure 8.7. A fair correlation between simulated and recorded water levels has been achieved in terms of:

• Peak water levels at Bulahdelah gauge – peak flood levels show a fair agreement for this event. The modelled level at peak is approximately 0.08 m lower and 0.17 m higher than the recorded levels for the first and second peaks respectively

• Peak water levels at the Crawford River gauge – peak flood levels show a relatively poor agreement for this event. The modelled level at peak is about 0.27-0.30 m lower than the maximum recorded level for the two peaks This difference in peak is due to the limited

rainfall data available across such a large catchment as well as limited bathymetric information along the Crawford River

• Catchment runoff response – the relative timing of the recorded and simulated water levels also shows a fair agreement in the overall volume and the behaviour of the storm

The results of the 18-24 March 2021 event therefore provide further confidence in the ability of the developed TUFLOW model to simulate actual catchment flood behaviour at Bulahdelah. Comparison with flood mark levels measured following the event

A number of flood marks were collected by MHL’s project team following the flood event. The results of the calibration were compared to the levels measured during the site visit. The comparison is provided in Figure E.7 of Appendix E. The model appears to reasonably represent the peak flood levels measured around Bulahdelah with level differences ranging between 0.04 and 0.23 m There are some larger differences near Markwell Bridge with difference of about 0.35-0.44 m.It is noted that while some of the flood marks in the Bulahdelah township showed clear lines along fences highlighting the peak flood level (Figure 8.12a&b), other flood marks where estimated based on visible stains and debris and may have a lower level of accuracy (Figure 8 12c&d).

Figure 8 11 Bulahdelah and Crawford River Water Levels Validation – 18-24 March 2021

(a) Flood Mark ID 7

(b) Flood Mark ID 8

(c) Flood Mark ID 1

(d) Flood Mark ID 5

Figure 8.12 Example of flood marks recorded following the March 2021 flood event

8.6 Discussion of Model Calibration and Validation

Despite some differences in observed and simulated peak water levels, the developed TUFLOW flood model appears to have performed reasonably well. This is supported by good comparisons between simulated and recorded water levels for both the calibration and validation events at Bulahdelah and Crawford River water level gauges, and agreement on catchment runoff response shown by flow hydrographs computed by TUFLOW and WBNM.

Some discrepancies between simulated and recorded water levels in Crawford River appear to be related primarily to limited availability of accurate rainfall information to determine the rainfall distribution over such a large catchment area. However, the use of rain radar information allowed the model to reproduce an overall good match in shape and volume of the various flood events and reasonable results were obtained when comparing the results to available flood marks and flood photographs.

To further assess the sensitivity of the flows from the Crawford catchment, additional model runs were undertaken by increasing the 5% AEP and 1% AEP flows from the Crawford catchment by 30%. It was found that this increase in flow generated an increase in flood level of approximately 0.11 m and 0.19 m at Bulahdelah and 0.21 m and 0.23 m at the Crawford gauge for the 5% AEP and 1% AEP, respectively. This is within the 500mm freeboard used for future planning purposes

Based on the results of the calibration and validation events, the model was therefore considered appropriate for the flood modelling.

9 Model sensitivity

A number of factors require some sensitivity analysis prior to completing the design runs. These factors include:

• Tailwater level: impact of reduced and increased water level at Bombah Broadwater was investigated.

• Losses: impact of reduced and increased losses were investigated.

• Roughness: impact of reduced and increased losses was investigated.

• Blockage: AR&R 2019 recommends running two blockage sensitivity scenarios including double design blockage and no blockage.

Sensitivity results are provided in Appendix F

9.1 Tailwater level sensitivity

In order to analyse the influence of the tailwater level on the flood behaviour, the following scenarios were modelled:

• Scenario with a 500 mm increase in tailwater level for the 1% AEP flood event.

• Scenario with a 500 mm decrease in tailwater level for the 1% AEP flood event.

Each of these scenarios was modelled for both the 45-minute and 18-hour critical durations. The following observations can be made:

• For the 18-hour critical duration event, the tailwater level has negligible impact on the flood level at the Bulahdelah Township and only increases the flood level at the downstream end of the catchment near Bombah Broadwater.

• For the 45-minute critical duration, the short duration does not allow the river to respond significantly to the rainfall and the tailwater has therefore a more direct change in level due to the low-lying floodplain. However, the majority of the additional flooding would occur at a lower level than for the long-duration flooding and remain predominantly along the river and vegetated areas.

9.2 Losses and roughness sensitivity analysis

In order to analyse the influence of losses and roughness on the flood behaviour, the following scenarios were modelled:

• No loss scenario with neither continuing losses nor initial losses in the pervious areas for the 1% AEP flood event.

• High loss scenario with 2.5 mm/hr continuing losses and 20 mm initial losses in the pervious areas for the 1%AEP flood event.

• Low roughness scenario with roughness reduced by 20% for the 1% AEP flood events.

• High roughness scenario with roughness increased by 20% for the 1% AEP flood events.

Each of these scenarios was modelled for both the 45-minute and 18-hour critical durations. The following observations can be made:

• The removal of all losses would generate increases in levels in the order of up to 0.100.15 m around both the township and the river catchment.

• High losses would result in decrease in flood level by up to 0.30-0.40 m within the river catchment and up to 0.05-0.10 m in the township catchment.

• Changes in material roughness by 20% impact river levels by up to +/- 0.15-0.25 m for the 1% AEP event around the river catchment for the 18-hour duration event but has minimal impact on the flood level for the 45-minute duration event

9.3 Blockage sensitivity analysis

In order to analyse the influence of blockage on the flood behaviour, the following scenarios were modelled:

• No blockage scenario for the 1% AEP event. These scenarios assumed that all the pits and pipes were free of blockages.

• Double design blockage scenario for the 1% AEP flood event. These scenarios consider the double design blockage for the cross-drainage structures as recommended by AR&R 2019 and the severe blockage value of 100% blockage for the pits and pipes system as described in AR&R 2019 Project 11.

Based on the results of this analysis, the following observations can be made:

• The no blockage scenario for the 18-hour duration event for the 1% AEP event results in local changes in flood level in the order of up to 0.10 m. Some areas would have reduced levels directly upstream of major culverts/pipes and some other areas would have increased levels downstream of major structures. There are a number of locations where there is a significant reduction in flood levels directly upstream of Bulahdelah Way and the Bulahdelah Bypass.

• The no blockage scenario for the 45-minute duration event for the 1% AEP event results in similar outcomes as the 18-hour duration. However, the increases in flood levels in the township are slightlyexacerbated with some local increasesalong drainage channels of up to approximately 0.30 m and further reduction in flood levels upstream of the Bulahdelah Bypass upstream of the township.

• The double blockage scenario for the 1% AEP event typically results in increases of up to 0.10 m in the township catchment with some local higher increase of up to 0.20-0.40 m along some of the main drainage channels and upstream of major culverts The 45minute duration event also highlights that the flooding would be diverted from the main drainage channels should there be high blockage of the various culverts that intersect these drainage channels at regular interval.

• In general, afflux is worse in the township under shorter duration events and upstream of major culverts where the local area acts as a detention basin (e.g. upstream of the Bulahdelah Bypass) for longer durations as a greater volume of water is unable to drain from this basin

10 Flood modelling results

10.1Flood modelling description

The TUFLOW hydraulic model was run for six (6) flood events including the 20%, 5%, 1%, 1 in 200 AEP, 1 in 500 AEP and PMF flood events. Two durations representing the upstream and downstream critical durations as described in Table 6.5 were modelled and an envelope of these two durations was produced to represent the flooding for each event.

10.2Flood mapping

10.2.1 Mapping filtering

The flood extents were filtered to remove shallow depths areas generated by the direct rainfall methodology. The criteria used to filter these shallow depths consisted of keeping areas where:

• Depth > 0.10 m; OR

• Depth > 0.05 m AND Velocity x Depth > 0.025 m2/s; OR

• Velocity > 2 m/s

Further to these criteria, “puddles” of less than 100 m2 were also removed from the extent. These filtering criteria are based on recent studies completed along the NSW coastline such as the Coastal Lagoons Catchments Overland Flood Study for Central Coast Council and the Racecourse Creek Flood Study and Option Assessment for MidCoast Council

10.2.2 Flood maps

Flood mapping presenting the peak flood level, peak flood depth and peak flood velocity envelop of each event is provided in Appendix G

10.3Flood levels outputs

The peak design flood levels at specific locations have been summarised in Table 10.1. The specific locations are illustrated in Figure 10.1.

10.4Comparison with previous study results

The outcome of the current study was compared to the results of the previous study provided by Council. The extent of the 1% AEP of the two studies as well as a long section of the Myall River over the area were both studies overlap for the 1% AEP flood event are provided in Figure 10 2 It can be noted that the flooding level at the township in the present study is lower than the previous study by up to 0.9 m and is approximately 1.5 m higher at the downstream end of the model. This is the result of the widespread flooding and associated storage not being represented in the 1D model downstream of the Bulahdelah Township. The crosssections of the previous model were about 500 m wide while the storage area downstream of the township can reach a width of nearly 3 km. This lack of storage generates a backwater effect increasing the water level significantly. It is also noted that significant storage occurs downstream of the extent of the previous model between the downstream end of the model and Bombah Broadwater.

Table 10.1 Summary of peak flood levels at various locations

Figure 10.2 Comparison of 1% AEP flood levels between current and previous flood study

11 Consequences on the community

11.1Preamble

The impact of flooding on the community is described in this section. The first step of understanding the impact of flooding on the community is to determine the flood behaviour within the catchment and identify key problem areas. Flood impact, road closure and flood damages can then be assessed, and more details are provided in this section.

11.2Flood behaviour

Flooding in the township catchment is ‘flashy’ in nature, with flood levels rising rapidly in response to relatively short durations of high intensity rainfall as opposed to extended periods of rainfall of lower intensity. This is confirmed by the critical duration ranging between 45 minutes and 2 hours as shown in Table 6.5. The river catchment has a slower response with a critical duration of 18 hours for all events except the PMF that has a 6-hour critical duration This was represented by the analysis of two critical durations as part of the design event modelling: a longer duration for the Myall River and a shorter duration for the flashy township catchment.

While there is some potential for rapid inundation of some properties and some roads in response to short durations of rainfall during rare events, the main flooding type in the study area is the flooding resulting from the river flooding. This flooding is slower and provides additional time to disseminate flood warnings Flood waters generally recede within hours or days after such an event

The study area contains various small drains flowing through the township that can be subject to overtopping during larger events.

Key issue areas potentially subject to deep flooding from the river include:

• Properties located west of Stroud Street and south of Jackson Street (including most properties along Jackson Street);

• Properties located in the two blocks enclosed within Prince Street, Richmond Street, Alexandra Street and Jackson Street;

• Properties east of Stroud Street and west of Crawford Street between Meade Street and Jackson Street; and

• The area around the intersection of River Street and Richmond Street

Key problem areas subject to shallower overland flooding include:

• Properties between Myall Street and Edgar Street at the southern end of the streets;

• Properties located along the drainage channel linking Stroud Street to Richmond Street via Alexandra Street;

• Properties along the flowpath linking Mackenzie Street to Crawford Street via Harold Street, Rosewood Street, Bulahdelah Way and Meade Street;

• Properties along the flowpath linking Bulahdelah Way to Crawford Street halfway

between Stuart Street and Blanch Street, as well as properties along the western side of Crawford Street between this flowpath and Stuart Street; and

• Properties at the northeastern end of Suncrest Close.

A number of these locations may be impacted by relatively frequent events like the 20% AEP.

11.3Key infrastructure assets

There are two main types of key infrastructure assets as presented below:

• The first type includes facilities that are occupied by emergency responders such as police stations, fire stations or SES Centres.

• The second type includes facilities with particularly vulnerable residents such as schools, childcare centres, aged care facilities and hospitals.

The locations of these key assets have been sourced from publicly available information (e.g. google map). A list of these facilities is provided in Table 11.1 along with a brief description of the flood affectation of each asset A map showing the location of these main structure is presented in Figure 11 1

Location

Police and Fire Stations

Bulahdelah Rural Fire Brigade

Bulahdelah Police Station

SES Centres

Comments on Flood Risk

The fire station is partly located within the PMF extent and access to the station is cut during PMF

Bulahdelah Police Station is outside the PMF extent but access to the station is cut during PMF.

SES There are no SES facilities located within the study area The nearest SES units are located at Stroud to the west, Forster to the east and Port Stephens to the south.

Hospital and Ambulance Stations

Bulahdelah Ambulance Station

Hospitals

The ambulance station is located within the PMF extent and access from the station may become limited during a PMF flood event.

There are no hospitals located within the study area. However, there is a nurse-led Community Health Centre known as the Bulahdelah Hospital that operates 7 am to 10.30 pm, 7 days per week. This centre is outside of the PMF extent. However, access to this venue by residents of various parts of Bulahdelah may be restricted as discussed in the following section.

Schools

St Joseph’s The school is outside of the PMF extent but access to school will be

Table 11 1 List of Key Infrastructure assets

Location Comments on Flood Risk

Primary School cut in such event.

Bulahdelah Central School

The school is outside of the PMF extent but access to school will be cut in such event.

Childcare Facilities and Preschools

Bulahdelah Preschool

The preschool is within the PMF extent and the property may be impacted by flood during a 1 in 200 AEP Access to the preschool would be cut during a PMF event

Aged Care Facilities and Retirement Villages

Great Lakes Nursing Home

This nursing home is partly located within the PMF extent and access to the station is cut during PMF NSW SES advised that this facility closed in 2022.

Designated Evacuation Centres

Bulahdelah Golf Club

The Bulahdelah Golf Club is used as an evacuation centre for the township of Bulahdelah. This venue is outside of the PMF extent. However, access to this venue by residents of various parts of Bulahdelah may be restricted as discussed in the following section.

11.4Road closure

An assessment of the frequency and hazard of road inundation is important for understanding the risk of vehicles becoming unstable, posing a risk to life for their drivers and passengers. It is also important for understanding evacuation risks, informing the classification of communities according to flood emergency response planning considerations. Measures to increase the flood immunityof critical roads could be considered as a resultofthis assessment.

Figure 11.2 depicts the flood events which result in road closures in the four lagoon catchments. Road closure was assumed as occurring when the depth of water over road reaches over 0.3 m, which is the depth that can start mobilising cars. Table 11.2 summarises the peak depth, duration of flooding over 0.3 m and time to depth above 0.3 m for each location presented on Figure 11.2

It can be noted from the flood mapping that a large number of locations get some water over the road very rapidly in the study area, although it may not become a significant problem every time. This is fairly common during overland flooding events.

Stroud Street, Jackson Street and Markwell Road near Markwell Bridge are inundated during a 20% AEP.

A few more roads become inundated from a 5% AEP in the township such as Stuart Street and Myall Street, as well as outside of the township such as the intersection of Frys Creek Road and Markwell Road, and a couple of locations along Bombah Point Road.

In the 1% AEP flood, several other roads become inundated at various locations around the township including Meade Street near Crawford Street, Prince Street, Alexandra Street and

River Street Markwell Road and Bombah Point Road become inundated closer to the township.MarkwellBackRoad, Suncrest Road and Myall Forest Road also become inundated. Crawford Street and Alexandra Street becomes cut at two locations during a 1 in 200 AEP event.

During a probable maximum flood event, numerous roads become cut around the town and major roads become inundated including Bulahdelah Way and the Pacific Highway. The above roads were considered inundated when the depth over the road exceeds 0.30 m as access would become impracticable.

Figure11.1

Keyinfrastructure locations

Legend

Studyarea

Extentof1%AEPflood

Keyinfrastructure

AgedCareFacilitiesand RetirementVillages

ChildcareFacilitiesand Preschools

DesignatedEvacuation Centres

FireStation

HospitalandAmbulance Stations

PoliceStation Schools

ReportMHL2789

ReviewofBulahdelah FloodplainRisk ManagementStudyandPlan

Golf

Table 11.2 Peak depth, duration of flooding over 0.3 m and time to depth above 0.3 m at road closure location

: Durations in italic are likely to be exceeded as the depth in the model was still higher than 0.15 m at the end of the simulation.

results

12.1Preamble

Once the flood mapping was completed for the main parameters (water level, depth and velocity), it became possible to determine the flood function, flood hazard and flood emergency response classification resulting from these data. Development of such categorisations is described in this section.

12.2Flood hazard

A starting point for the assessment of Flood Life Hazard categories is to better understand the flood hazard. National Best Practice Guidelines present a set of hazard vulnerability curves shown in Figure 12.1.This shows how flood depths, velocities and depth-velocity product affect the stability of vehicles, pedestrians and buildings.

Figure 12.1 General flood hazard vulnerability curves

Source: AIDR (2017)

The above hazard vulnerability categories have been mapped for the 1% AEP and the PMF events for the entire study area and are presented in Appendix H

During a 1% AEP flood event, the extent of hazard conditions between H3 and H6 are typically concentrated along main rivers and drainage channels, while the majority of the township flooding is classified as H1 or H2 hazard category. However, some larger hazard ranging between H3 and H5 can be observed in the area west of Stroud Street between Meade Street and Jackson Street as well as between Prince Street and Alexandra Street south of Richmond Street. The Myall River generates a small secondary branch inundating the properties along the southern half of River Street, south of Richmond Street and east of Myall Street with hazard categories between H3 and H4. A large part of the floodplain is classified as H5 category. During a PMF flood event, the majority of the township would be subject to significant hazards of H3 or above category with only a few elevated areas in the northwestern corner, the southeastern corner and the northeastern corner ofthe township remaining eithernot impacted or subject to lower flooding categories (H1 and H2) Most properties located along Frys Creek and Markwell Back Road would also be subject to H3 or above hazard categories.

12.3Flood function (hydraulic categorisation)

Hydraulic categorisation is a useful tool in assessing the suitability of land use and development in flood-prone areas. The Australian Disaster Resilience Handbook (AIDR, 2017) describes the following three hydraulic categories of flood-prone land:

• Floodway / Flow Conveyance – Flow conveyance areas are defined as those areas where a significant flow of water occurs. They typically flow continuously from the upper reaches of waterways and flowpaths within the catchment to the outlet during a flood. These flows often align with naturally defined channels. Theyare areasthat, even if onlypartiallyblocked by changes in topography or development, cause a significant redistribution of flood flow or a significant increase in flood levels. They are often, but not necessarily, areas of deeper flow or areas where higher velocities occur. In the DFE, they generally extend beyond the waterway banks.

• Flood Storage – During a flood event, significant amounts of floodwater can also extend into, and be temporarily stored in, areas of the floodplain. This water flows downstream as the flood recedes. Where storage is important in attenuating downstream flood flows and levels, areas storing this water are classified as flood storage areas. Filling of flood storage areas reduces their ability to attenuate downstream flood flows and, as a result, flood flows and flood levels may increase.

• Flood Fringe – Flood-fringe areas make up the remainder of the flood extent for the particular event. It is the area where the effects on flood function are not a constraint. Developing in flood-fringe areas is unlikely to significantly alter flood behaviour, beyond the broader impact of changes to run-off because of urbanisation within the catchment. However, other flood-related constraints may exist in flood-fringe areas.

These qualitative descriptions do not prescribe specific thresholds for determining the hydraulic categories in terms of model outputs, and such definitions may vary between floodplains depending on flood behaviour and associated impacts. For the purposes of the Review of Bulahdelah FRMSP, hydraulic categories have been defined as per the criteria in Table 12 1. These criteria have recently been used on the Racecourse Creek catchment as well as other catchments in the Central Coast such as the Coastal Lagoons Catchments

© Crown 2021

Classification: Release by consent

Overland Flood Study completed in 2020. The floodway criterion has been selected as it provides improved continuity of flow along the various flow paths and considers areas of deeper flows. The flood storage criteria were selected as they have been commonly applied on various recent overland studies around NSW and considers areas with deep flood depth allowing storage of flood water.

Hydraulic category mapping for the 1%AEP and PMF design events ispresented in Appendix H

Table 12 1 Hydraulic category criteria

Hydraulic Category Criteria

Floodway Velocity x Depth > 0.25 m2/s

Description

Flowpaths and channels where a significant proportion of flood flows are conveyed

Flood Storage Depth > 0.5 m, Not Floodway Areas that temporarily store floodwaters and attenuate flood flows

Flood Fringe Depth < 0.5 m, Not Floodway or Flood Storage

Generally shallow, low velocity areas within the floodplain that have little influence on flood behaviour

During a 1% AEP flood event, the floodways typically remain within the main river and drainage channels A small anabranch branches off the Myall River and cuts through some properties at the southern end of River Street and western end of Richmond Street. A larger anabranch shortcuts the bend of the Myall River located at the junction with the Crawford River and flows through the western end of Jackson Street, the southern end of Prince Street and along the western side of the River Myall Holiday Resort.

The properties located west of Crawford Street between Meade Street and Jackson Street as well as between Prince Street and Alexandra Street South of Richmond Street are located within a flood storage area. During a 1% AEP flood event, the larger part of the floodplain is either classified as floodway or flood storage.

During a PMF flood event, the southwestern half of the township would be located with the floodway of the river. Properties located along drainage channels flowing through the township would also be within the floodway. A fewmore propertiesaround the township would be located within the flood storage area. Mostproperties along Frys Creekand Markwell Back Road would also be located within the floodway. During a PMF flood event, the larger part of the floodplain would be classified as floodway.

13 Conclusion

The flood study review has been completed to provide a detailed flooding assessment of Bulahdelah and the surrounding local catchments The previous studies, while providing relevant information that relates to the flooding of Bulahdelah, were completed prior to the creation of the latest guidelines The objective of this study is therefore to improve understanding of flood behaviour and impacts, and better inform management of flood risk in the study area. The study also provides a sound technical basis for any further flood risk management investigation in the area.

The key components of the flooding assessment included:

• Review of previous studies and available data

• Community consultation

• Hydrological analysis and modelling

• Hydraulic analysis and modelling

• Calibration and validation of the hydraulic model

• Sensitivity analysis

• Flood mapping

• Description of consequences of flooding

• Development of a draft flood study review report followed by a final report

The flood maps appended to this report are presenting the flood levels, depths and velocities for the critical duration and rainfall pattern of a full set of events including the 20%, 5%, 1%, 1 in 200, 1 in 500 AEP and PMF events and represent an envelope of the critical duration/pattern of a selected representative catchment for the Bulahdelah township and the critical duration/pattern for the Myall River. The township catchment is very reactive with very short critical durations between 45 minutes and 2 hours to reach the peak level, while the river catchment has typical critical durations ranging between 6 hours and 18 hours

Sensitivity analysis highlighted the following points:

• Tailwater conditions (including sea level rise) typically have relatively small impact on river flooding levels due to a number of natural choke points located at various locations along the Myall River raising the level much higher than the downstream level However, for smaller catchment events, that do not lead to a major response in the overall river catchment flooding, changes in Bombah Broadwater levels will propagate directly up the Myall River.

• Changes in roughness or in antecedent conditions of the catchment (wet/dry catchment leading to varying losses) could have moderate impact on river flooding and low impacts on overland flooding

• Blockages of structures can have severe impact in areas with no gravity flow that only relies on the drainage network (e.g. ponding area); maintaining the pits and pipes network is essential to avoid exacerbating the flooding in those locations. It is also

noted that no blockage can lead to small increases in flood level at the downstream end of the township.

The above results allowed the definition of the flood hazard (i.e. H1-H6 flood life hazard categories) and hydraulic categories in the Bulahdelah catchment. These have been created and mapped to inform development control planning.

Results of the model allow the identification of main flooding areas, key infrastructure assets impacted by flooding, and road closures around the catchments. Key infrastructure typically may have access issues during severe flood events rather than flooding issues, except during the PMF event.

Apart from Stroud Street, road closures predominantly occur on secondary roads with most of the major road closures occurring for the PMF only. It is also noted that given the flashy behaviour of the catchment, flooding and road closures in the upper township catchment would be of relatively short duration while flooding of the areas surrounding the river may last from several hours to a few days

© Crown 2021

Classification: Release by consent

References

Australian Government (Attorney – General’s Department). (2009). Flood Preparedness (Manual 20). Australian Emergency Manuals Series.

Ball, J. E. (2019). Australian Rainfall and Runoff - A guide to flood estimation. Commonwealth of Australia (Geoscience Australia).

BMT. (2015). Lower Myall River and Myall Lake Flood Study. Great Lake Shire Council.

Bureau of Meteorology. (2003). The Estimate of Probable Maximum Precipitation in Australia: Generalised Short Duration Method.

Bureau of Meteorology. (2003). The Estimation of Probable Maximum Precipitation in Australia: Generalised Short-Duration Method.

Bureau of Meteorology. (2005). Guidebook to the Estimation of Probable Maximum Precipitation Generalised Tropical Storm Method.

Bureau of Meteorology. (2006). Guidebook to the Estimation of Probable Maximum Precipitation: Generalised Southeast Australia Method.

Department of Infrastructure, Planning and Natural Resources. (2003). Myall River Floodplain Risk Management Plan for Bulahdelah. Great Lakes Council.

Department of Land and Water Conservation. (2002). Myall River Floodplain Risk Management Study for Bulahdelah. Great Lakes Council.

NSW Fisheries Office of Conservation. (2003). Monitoring changes in Crawford River fish community following replacement of an ineffective fishway with a vertical slot fishway design: Results of an eight year monitoring program.

Office of Environment and Heritage. (2019). Floodplain Risk Management Guide, Incorporatng 2016 Australian Rainfall and Runoff in Studies. Office of Environment and Heritage.

Pilgrim, D. H. (1987). Australian rainfall and runoff - A Guide to Flood Estimation. Institution of Engineers, Australia. Public Works. (1991). Bulahdelah Flood Appraisal. Great Lakes Shire Council. Public Works. (1994). Frys Creek Flood Study. Great Lakes Shire Council.

Survey marks checks

Distribution of differences between Survey Marks and LiDAR elevation

Floor level survey results

Community Consultation Community Survey Form and Information Pack

Community questionnaire results summary

Key concerns and comments from the community questionnaires

ARR 2019 Data Hub Results for Bulahdelah

Results - ARR Data Hub

[STARTTXT]

Input Data Information

[INPUTDATA]

Latitude,-32.351400

Longitude,152.160000

[END_INPUTDATA]

River Region [RIVREG]

Division,South East Coast (NSW)

River Number,9

River Name,Karuah River

[RIVREG_META]

Time Accessed,30 November 2020 02:14PM

Version,2016_v1

[END_RIVREG]

ARF Parameters [LONGARF]

Zone,SE Coast

a,0.06

b,0.361

c,0.0

d,0.317

e,8.11e-05

f,0.651

[LONGARF_META]

Time Accessed,30 November 2020 02:14PM

Version,2016_v1

[END_LONGARF]

Storm Losses [LOSSES]

ID,18945.0

Storm Initial Losses (mm),45.0

Storm Continuing Losses (mm/h),3.1

[LOSSES_META]

Time Accessed,30 November 2020 02:14PM

Version,2016_v1

[END_LOSSES]

Temporal Patterns [TP]

code,ECsouth

Label,East Coast South

[TP_META]

Time Accessed,30 November 2020 02:14PM

Version,2016_v2

[END_TP]

Areal Temporal Patterns [ATP]

code,ECsouth arealabel,East Coast South [ATP_META]

Time Accessed,30 November 2020 02:14PM

Version,2016_v2

(0.002),0.6 (0.002),0.8 (0.003),4.6 (0.013),7.5 (0.019)

4320 (72.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),1.8 (0.004),3.2 (0.007)

[PREBURST_META]

Time Accessed,30 November 2020 02:14PM

Version,2018_v1

Note,Preburst interpolation methods for catchment wide preburst has been slightly altered. Point values remain unchanged.

[END_PREBURST]

10% Preburst Depths

[PREBURST10]

min (h)\AEP(%),50,20,10,5,2,1

60 (1.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

90 (1.5),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

120 (2.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

180 (3.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

360 (6.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

720 (12.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

1080 (18.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

1440 (24.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

2160 (36.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

2880 (48.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

4320 (72.0),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000),0.0 (0.000)

[PREBURST10_META]

Time Accessed,30 November 2020 02:14PM

Version,2018_v1

Note,Preburst interpolation methods for catchment wide preburst has been slightly altered. Point values remain unchanged.

[END_PREBURST10]

25% Preburst Depths [PREBURST25]

min (h)\AEP(%),50,20,10,5,2,1

60 (1.0),0.0 (0.000),0.0 (0.000),0.0

Time Accessed,30 November 2020 02:14PM

Version,2018_v1

Note,Preburst interpolation methods for catchment wide preburst has been slightly altered. Point values remain unchanged.

[END_PREBURST25]

75% Preburst Depths [PREBURST75]

Note,Preburst interpolation methods for catchment wide preburst has been slightly altered. Point values remain unchanged.

[PREBURST90_META]

Time Accessed,30 November 2020 02:14PM

Version,2018_v1

Note,Preburst interpolation methods for catchment wide preburst has been slightly altered. Point values remain unchanged.

[END_PREBURST90]

Interim Climate Change Factors [CCF]

,RCP 4.5,RCP6,RCP 8.5

2030,0.869 (4.3%),0.783 (3.9%),0.983 (4.9%)

2040,1.057 (5.3%),1.014 (5.1%),1.349 (6.8%)

2050,1.272 (6.4%),1.236 (6.2%),1.773 (9.0%)

2060,1.488 (7.5%),1.458 (7.4%),2.237 (11.5%)

2070,1.676 (8.5%),1.691 (8.6%),2.722 (14.2%)

2080,1.810 (9.2%),1.944 (9.9%),3.209 (16.9%)

2090,1.862 (9.5%),2.227 (11.5%),3.679 (19.7%)

[CCF_META]

Time Accessed,30 November 2020 02:14PM

Version,2019_v1

Note,ARR recommends the use of RCP4.5 and RCP 8.5 values. These have been updated to the values that can be found on the climate change in Australia website.

[END_CCF]

Probability Neutral Burst Initial Loss

[BURSTIL]

min (h)\AEP(%),50.0,20.0,10.0,5.0,2.0,1.0

60 (1.0),29.3,15.7,14.1,13.7,14.2,12.3

90 (1.5),30.9,18.5,17.6,17.3,14.4,8.8

120 (2.0),27.8,16.9,15.8,16.1,15.0,11.7

180 (3.0),30.2,18.9,16.1,14.8,12.8,9.2

360 (6.0),24.0,16.4,15.4,13.0,11.5,6.1

720 (12.0),25.1,18.4,16.7,16.6,16.9,9.5

1080 (18.0),29.1,21.8,20.4,19.5,18.4,8.9

1440 (24.0),34.8,24.9,23.3,22.0,22.7,8.0

2160 (36.0),36.9,27.2,26.5,24.3,25.3,9.4

2880 (48.0),42.5,33.9,33.1,33.7,26.1,10.6

4320 (72.0),47.2,40.6,39.4,42.8,33.7,18.2

[BURSTIL_META]

Time Accessed,30 November 2020 02:14PM

Version,2018_v1

Note,As this point is in NSW the advice provided on losses and pre-burst on the <a href="./nsw_specific">NSW Specific Tab of the ARR Data Hub</a> is to be considered. In NSW losses are derived considering a hierarchy of approaches depending on the available loss information. Probability neutral burst initial loss values for NSW are to be used in place of the standard initial loss and pre-burst as per the losses hierarchy.

[END_BURSTIL]Transformational Pre-burst Rainfall

[PREBURST_TRANS]

min (h)\AEP(%),50.0,20.0,10.0,5.0,2.0,1.0

60 (1.0),16.1,29.7,31.3,31.7,31.2,33.1

© Crown 2021

90 (1.5),14.5,26.9,27.8,28.1,31.0,36.6

120 (2.0),17.6,28.5,29.6,29.3,30.4,33.7

180 (3.0),15.2,26.5,29.3,30.6,32.6,36.2

360 (6.0),21.4,29.0,30.0,32.4,33.9,39.3

720 (12.0),20.3,27.0,28.7,28.8,28.5,35.9

1080 (18.0),16.3,23.6,25.0,25.9,27.0,36.5

1440 (24.0),10.6,20.5,22.1,23.4,22.7,37.4

2160 (36.0),8.5,18.2,18.9,21.1,20.1,36.0

2880 (48.0),2.9,11.5,12.3,11.7,19.3,34.8

4320 (72.0),0.0,4.8,6.0,2.6,11.7,27.2

[PREBURST_TRANS_META]

The tranformational pre-burst is intended for software suppliers in the NSW area and is simply the Initial Loss - Burst Initial Loss. It is not appropriate to use these values if considering a calibrated initial loss.

[END_PREBURST_TRANS]

[ENDTXT]

MHL2789 – D4

Classification: Release by consent

Calibration data and results

Marker intensity timeseries between 4pm on 28-02-2013 and 12am on 02-03-2013 at:

Upper Myall River Catchment (32.192°S, 152.118°E) – UMR

Marker intensity timeseries between 6pm on 20-04-2015 and 6am on 21-04-2015 at:

Upper Myall Catchment (32.192°S, 152.118°E) – UMC

Bulahdelah Bridge Gauge (32.413°S, 152.209°E) – BBG

Bulahdelah Bridge Gauge (32.413°S, 152.209°E) – BBG

Upper Crawford River Catchment (32.391°S, 152.128°E) – UCR

Upper Crawford River Catchment (32.391°S, 152.128°E) – UCR

Figure E.1 Rainfall intensity summary at key location for March 2013

Figure E.2 Rainfall intensity summary at key location for April 2015 storm

UCR UCR BBG BBG

Marker intensity timeseries between 12am on 18-03-2021 and 12am on 25-03-2021 at:

Upper Myall River Catchment (32.192°S, 152.118°E) – UMR

Bulahdelah Bridge Gauge (32.413°S, 152.209°E) – BBG

Upper Crawford River Catchment (32.391°S, 152.128°E) – UCR

Figure E.3 Rainfall intensity summary at key location for March 2021

UCR BBG

Radar data unavailable between 10.30 pm on 28-02-2013 and 1.30 pm on 01-03-2013

Figure E 4 – Rain radar at 1-hour interval between 4pm on 28-02-2013 and 10pm on 02-03-2013 from TheWeatherChaser.com website

Figure E.4 – Newcastle rain radar at 1-hour interval between 4pm on 28-02-2013 and 10pm on 02-03-2013 from TheWeatherChaser.com website (continued)

Figure E 5 – Newcastle rain radar at 30-minute interval between 2pm on 20-04-2015 and 5.30pm on 21-04-2015 from TheWeatherChaser.com website

Figure E.5 – Newcastle rain radar at 30-minute interval between 2pm on 20-04-2015 and 5.30pm on 21-04-2015 from TheWeatherChaser.com website (continued)

Figure E.6 – Newcastle rain radar at 2-hour interval between 8pm on 18-03-2021 and 6.00pm on 22-03-2021 from TheWeatherChaser.com website

Figure E.6 – Newcastle rain radar at 2-hour interval between 8pm on 18-03-2021 and 6.00pm on 22-03-2021 from TheWeatherChaser.com website (continued)

Comments:

Calibration Result

Table E.1 April 2015 validation event photographs and model results comparison (photographs courtesy of Eric Saville; aerial source: Nearmap 2021)

Jackson Street looking west near intersection with Alexandra Street showing deep flood water across roadway.

Comments: Jackson Street looking west near intersection with Alexandra Street showing edge of flooding.

Comments:

Looking east across Stroud Street from Jackson Street showing flowpath between Stroud Street and Crawford Street

Comments:

Looking south down Stroud Street from Jackson Street showing extent of deep flooding across Stroud Street

Comments:

Looking east on property along Jackson Street showing extent of flooding at the back of properties.

Comments:

Looking south onto open area flooding west of Stroud Street

Figure E.7 March 2021 validation event flood marks measurements and model results comparison

Model sensitivity results

Flood maps

Flood hazard and flood function maps

110B King Street

Manly Vale NSW 2093