keane road strategic link study to public review...keane wetland for the krsl was undertaken, based...

Keane Road Strategic LinkHydrologic Study – Response to Public Environmental Review

Submissions

March 2014

EnviroWorks Consulting & City of Armadale 2815‐01R02v03

3304‐01 / R02 v03 ii

DOCUMENT STATUS

Version Doc type

Reviewed by Approved by

Distributed to Date issued

V01 Draft CLA CLA Laura Todd, City of Armadale 07/03/2014

V02 Final Laura Todd CLA Laura Todd, City of Armadale 16/03/2014

V03 Final Laura Todd CLA Laura Todd, City of Armadale 18/03/2013

PROJECT DETAILS

Project Name 2815‐01R02v03

Client EnviroWorks Consulting & City of Armadale

Client Project Manager Laura Todd

Water Technology Project Manager Christine Lauchlan Arrowsmith

Report Authors Tristan Graham, Christine Lauchlan Arrowsmith

Job Number 3304‐01

Report Number R02

Document Name 3304‐01_Report‐R02V03

Cover Photo:

Copyright

Water Technology Pty Ltd has produced this document in accordance with instructions from EnviroWorks Consulting & City of Armadale for their use only. The concepts and information contained in this document are the copyright of Water Technology Pty Ltd. Use or copying of this document in whole or in part without written permission of Water Technology Pty Ltd constitutes an infringement of copyright.

Water Technology Pty Ltd does not warrant this document is definitive nor free from error and does not accept liability for any loss caused, or arising from, reliance upon the information provided herein.

15 Business Park Drive

Notting Hill VIC 3168

Telephone (03) 8526 0800

Fax (03) 9558 9365

ACN No. 093 377 283

ABN No. 60 093 377 283


3304‐01 / R02 v03 iii

EXECUTIVE SUMMARY

Water Technology was commissioned by EnviroWorks Consulting and the City of Armadale to undertake further local scale hydrologic assessment of the Keane Road Strategic Link (KRSL), in response to submissions received on the KRSL Public Environmental Review (PER).

Submissions on the KRSL PER have raised concerns that the proposal may not meet the Environmental Protection Authority (EPA) objective for hydrological processes as follows:

To maintain the hydrological regimes of groundwater and surface water so that existing and potential uses, including ecosystem maintenance, are protected.

Specifically, concerns have been raised that the road may interrupt hydrological processes and this could have implications on the ecology of the area. Concerns have also been raised regarding the robustness and verification of the previous hydrological modelling (Water Technology, 2013).

As a result of these concerns raised in submissions, further hydrological assessment and model verification has been undertaken in this report, in order to confirm the predicted hydrological impacts and recommended management measures for the KRSL project.

Hydrology Model Validation

On the basis of validation steps described within this report, as summarised below, it is concluded that the surface water model produced by Water Technology for the KRSL project is robust, and accurately represents the hydrology of the study area including the proposed KRSL alignment. Specifically model validation steps have included:

Confirmation of hydrological assumptions used within the model and as the basis for recommendations.

Re‐running the model using finer scale topography (1 m Digital Elevation Model sourced from the Department of Water).

Ground truthing the model via site inspection and aerial photography interpretation.

Each of these steps is described further below.

Confirmation of Hydrological Assumptions

Hydrological assumptions which have been confirmed as appropriate are summarised below:

The Anstey‐Keane wetland area is classified as a “dampland” wetland system in the Wetland Atlas for the Swan Coastal Plain (Hill et al, 1996). Management objectives for a dampland are consistent with recommendations made by Water Technology and the EPA objective for hydrological processes.

A series of 15 soil bores were drilled by Douglas Partners in November 2008 (Appendix D of the PER) which show the presence of fine to medium grained sand (Bassendean Sand) along the full length of the alignment, with a minimum thickness of 0.6 m. Although the soil conditions show the presence of a highly permeable surface sand layer, the modelling assumption adopted is for no initial loss of rainfall to groundwater. This is conservative, as it assumes a fully saturated groundwater condition and therefore does not include any infiltration of rainfall into the sand layer.

JDA (2012) describes the groundwater system underlying this area as an unconfined aquifer containing generally fresh to slightly brackish groundwater, with slightly acid to neutral pH. The water table is shallow, rising to within 0.5 m of the surface during winter, depending on surface elevation. The seasonal variation of the water table is about 2m, reaching a maximum elevation in September/October and a minimum elevation in April/May. These groundwater conditions were accurately incorporated into the Water Technology (2013) modelling as follows:


3304‐01 / R02 v03 iv

o The application of an approach for assessing the loss of rainfall to groundwater during a storm event (i.e. design losses) which is consistent with previous studies in this area (e.g. JDA, 2002) and the approach recommended in Australian Rainfall and Runoff (1998)

o The assumption of no initial losses to simulate an initially wet catchment, which is analogous to winter conditions with a high groundwater elevation.

Use of More Detailed Topography

Updated two‐dimensional (2D) rainfall on grid modelling of surface water systems across the Anstey‐Keane Wetland for the KRSL was undertaken, based on the 1 m Digital Elevation Model sourced from the Department of Water. This provides significantly higher resolution of the local topography in the hydraulic model compared to the previous investigation (Water Technology, 2013). The results of the updated assessment are consistent with previous findings (Water Technology, 2013) and have indicated the following:

Based on flow paths identified, it is concluded that the dampland is currently highly modified by unsealed roads/tracks which have become un‐natural man made drainage channels. These unsealed roads/tracks currently cause un‐natural surface water blockage, ponding and channelized flow in areas of the dampland which would have previously received natural sheet flow (diffuse flow) prior to human modification via the construction of unsealed tracks and ongoing use of these tracks for off road driving.

There are two ‘overland’ locations along the proposed KRSL road alignment where surface flows and ponding currently exist, which are associated with vehicle tracks, collecting and directing flows from upstream (Locations A and B). The flow depths and extents in these locations are limited.

For all rainfall events ponding of surface runoff occurs north of the intersection of Anstey Road and Keane Road, which is associated with Baileys Branch Drain (Location C). The extent of ponding in this area is dependent on the capacity of the drain itself. The water surface elevation and ponding extent shown is consistent with previous government modelling (Department of Water, 2009).

Ground Truthing

In addition to re‐running the model using finer scale topography (1 m DEM), a further validation step taken as part of this report has been ground truthing of the model as follows:

Conducting a site inspection to confirm the locations of topographic low points and existing artificial disturbances (such as unsealed tracks) causing channelized flow (Figure 3‐3);

Comparing surface water inundation clearly visible in high resolution aerial photography after recent rainfall with model outputs.

A GIS based review of the high resolution aerial imagery against the model results has found that the surface water distribution across the study area clearly visible in the aerial image is well represented in the model results.

Impact Assessment and Recommendation of Management Measures

To ensure the KRSL has no negative impact on the existing surface water systems (and where possible improves the existing degraded surface water systems) it is necessary to:

Provide continued hydrological connectivity for the three flow paths / areas of ponding identified which coincide with the proposed KRSL alignment (locations A, B, and C).

Ensure that existing ponding currently occurring upstream of the proposed KRSL alignment is not increased or decreased;


3304‐01 / R02 v03 v

Where possible, create diffuse flow (sheet flow) downstream of the KRSL alignment, in locations where current channelized flow is occurring un‐naturally down artificial unsealed roads/tracks.

To achieve this, the following management measures are recommended for three locations (locations A, B and C):

Provision of a series of aligned and adjacent small culverts or ‘small culvert arrays’ directing the flows horizontally underneath the road to ensure that:

o Diffuse flow (sheet flow) is created downstream (i.e. channelised flow does not occur downstream of the road);

o Ponding is not increased or decreased upstream (the small culvert array will allow water to flow underneath the road in an equivalent manner in terms of flow rates and volumes to pre‐development)

Appropriate design of the outflow of the culverts (such as rock rip rap, swales and riffle zones) is recommended to ensure diffuse flow is created downstream.

The recommended design features a series of 300mm diameter pipe culverts located at each of the identified flow paths / areas of ponding (locations A, B and C). Each array of culverts maintains the hydrologic connectivity and ensures peak 100 year flow rates are maintained without overtopping of the proposed KRSL roadway.

Recommendations for locations A and B are similar to those which were previously proposed (Water Technology, 2013) involving six 300mm diameter pipe culverts installed in an array, to maintain existing hydraulic connectivity and produce downstream “diffuse” flow in these locations.

Location C is near the existing Bailey Branch Drain which currently passes through 2 x 750mm diameter culverts at Keane Road. These culverts have a capacity of around 2.6m3/s. As detailed in previous government modelling (Department of Water, 2009) and shown in the current model results, ponding of water does already occur upstream of the proposed KRSL, north of the intersection of Keane Rd and Anstey Rd (in private farmland) under existing conditions for the 100 year ARI event. This existing upstream ponding within private farmland is connected to ponding downstream of the proposed KRSL alignment within native vegetation. Therefore instead of changing the capacity of the drain (which may change the current hydrology including ponding) it is proposed to provide a series of six 300mm diameter pipe culverts underneath KRSL near Bailey’s Branch drain, to maintain existing hydrological conditions including ponding on both sides of the KRSL under the 100 year ARI event.

Summary

The results of this assessment are consistent with the previous hydrological assessment (Water Technology, 2013) as follows:

There are three overland flow paths / areas of ponding (locations A, B and C) across the proposed KRSL alignment.

One artificial channelised flow path also exists across the proposed KRSL alignment – Bailey’s Branch Drain (termed flow path D in the 2013 Water Technology report).

A series of 300m diameter pipe culverts provided at locations A, B, and C will ensure the connectivity of these flow paths / areas of ponding across the proposed KRSL alignment.

The inclusion of an adequate number of appropriately sized culverts at the three locations, as identified in this report, will ensure there are no impacts on the local surface water system.

There are no impacts of the road on the local groundwater or surface water system as the road design allows for the provision of culverts to maintain flow connectivity; the un‐kerbed


3304‐01 / R02 v03 vi

profile with swale arrangement facilitates infiltration; and the use of the swale for water quality treatment along the alignment will minimise water quality impacts.

As there are no impacts from the road to surface water flow paths and no impacts on groundwater level, flows or water quality, it is unlikely there would be impacts to groundwater dependent or surface water dependent vegetation or ecosystems.

On the basis of the City of Armadale’s commitment to install the recommended management measures (water connectivity culverts) it is concluded the project can meet the EPA objective for Hydrological Processes:



3304‐01 / R02 v03 vii

TABLE OF CONTENTS

1. Introduction .................................................................................................................... 1

1.1 Background .............................................................................................................................. 1

2. Local Hydrology and Confirmation of modelling and management assumptions .............. 1

2.1 Wetland Classification ............................................................................................................. 1

2.1.1 Dampland Management Objectives ........................................................................................ 2

2.2 Soils and Modelling Assumptions ............................................................................................ 2

2.3 Groundwater Conditions, Levels and Modelling Assumptions ............................................... 5

3. Revised Surface Water Modelling .................................................................................... 6

3.1 Purpose .................................................................................................................................... 6

3.2 Finer Scale Topography Model Update Overview ................................................................... 6

3.3 Inflows ..................................................................................................................................... 7

3.4 Topography .............................................................................................................................. 7

3.5 Flow Paths ............................................................................................................................... 9

3.5.1 Flow Path Delineation ............................................................................................................. 9

3.5.2 Modelled Flow Paths ............................................................................................................. 12

3.5.3 Impacts of External Inflows ................................................................................................... 17

3.5.4 Recommendations to Maintain Hydrological Connectivity and Prevent Increased Ponding 20

4. Conclusions .................................................................................................................... 22

4.1 Local Hydrology ..................................................................................................................... 22

4.2 Revised Surface Water Modelling ......................................................................................... 22

5. References ..................................................................................................................... 25

LIST OF FIGURES

Figure 2‐1 Wetland Classification (Hill et al, 1996) ........................................................................ 1 Figure 2‐2 Surface Geology Classification (Perth Groundwater Atlas, 2004) ................................. 3 Figure 2‐3 Surface soils stratigraphy and groundwater levels based on borelog data .................. 4 Figure 2‐4 Photo looking along the Keane Road alignment towards Skeet Road .......................... 4 Figure 2‐5 Groundwater Contours across the Keane Road site (dark blue = maximum

groundwater elevation, light blue = groundwater elevation May 2003; Perth Groundwater Atlas http://www.water.wa.gov.au/idelve/gwa/ accessed 04/03/14) .. 5

Figure 3‐1 Site Topography ............................................................................................................ 8 Figure 3‐2 Flow Path Delineation ................................................................................................. 10 Figure 3‐3 Aerial image of the Keane Anstley Dampland including local site photos .................. 11 Figure 3‐4 Maximum water depths for the 100 year ARI rainfall event 9 hour critical duration 14 Figure 3‐5 Maximum water depths for the 10 year ARI rainfall event 9 hour critical duration .. 15 Figure 3‐6 Water depths for the 10 year ARI critical duration event compared to aerial imagery

from September 2013 ................................................................................................. 16 Figure 3‐7 Rainfall Depths at Armadale (9001) for 2013 .............................................................. 17 Figure 3‐8 Baileys Branch Drain Existing System Details (Figure 10c reproduced from FMDADS,

Department of Water, 2009)....................................................................................... 19 Figure 3‐9 100 year ARI event results showing location of proposed culverts on the KRSL to

ensure flow connectivity ............................................................................................. 21


3304‐01 / R02 v03 1

1. INTRODUCTION

1.1 Background

Water Technology has been commissioned by EnviroWorks Consulting and the City of Armadale to undertake further local scale hydrologic assessment of the Keane Road Strategic Link (KRSL), in response to submissions received on the KRSL Public Environmental Review (PER).

This report provides hydrological information in response to the submissions.

2. LOCAL HYDROLOGY AND CONFIRMATION OF MODELLING AND MANAGEMENT ASSUMPTIONS

2.1 Wetland Classification

The Anstey‐Keane wetland area is classified as a “dampland” wetland system in the Wetland Atlas for the Swan Coastal Plain (Hill et al, 1996), as shown in Figure 2‐1. A dampland is defined in the Atlas as a seasonally waterlogged basin, based on the geomorphic classification of wetlands for the Perth‐Bunbury region (Semeniuk & Semeniuk, 1995)

Figure 2‐1 Wetland Classification (Hill et al, 1996)

For such a wetland system, the soils/substrate is saturated with water, but the water does not inundate the soil surface across the majority of the wetland (at their most wet under prevailing conditions). They are saturated to the extent that they develop wetland characteristics, such as wetland soils, wetland plants, and distinct communities from surrounding dryland. The water is present in between sediments as interstitial waters, also known as sediment pore waters (Department of Environment and Conservation, 2012). In the case of the Keane Road wetland complex the wetlands are seasonally waterlogged.


3304‐01 / R02 v03 2

2.1.1 Dampland Management Objectives

Water Technology’s recommended management objectives for this dampland type of wetland include:

Avoid impacts to surface water or groundwater flows, levels, volumes and quality.

Where existing degradation and modification of the dampland has already occurred, restore wetland characteristics and processes to pre‐degradation state as far as practicable.

The following Environmental Protection Authority (EPA) objectives are also relevant:

Hydrological Processes EPA Objective: To maintain the hydrological regimes of groundwater and surface water so that existing and potential uses, including ecosystem maintenance, are protected.

Inland Waters Environmental Quality EPA Objective: To maintain the quality of groundwater and surface water, sediment and biota so that the environmental values, both ecological and social, are protected.

The above management objectives have formed the basis for recommendations made in this report.

2.2 Soils and Modelling Assumptions

The Perth Groundwater Atlas (2004) classifies provides a classification of the surface geology across the Anstey‐Keane Wetland site. This classification is derived from the Department of Industry and Resources (DOIR) 250k Geology dataset. The geological units were re‐classified based on groundwater significance. The set was generated from ‘Perth Region Aquifer Modelling System’ project and published in Davidson, W.A. and Yu, X. (2005).

The classification for the site is provided in Figure 2‐2. There are three soil classes relevant to the Keane Road alignment:

Bassendean Sand: quartz sand (dunes)

Swamp and Lacustrine deposits: peat, peaty sand and clay,

Guildford Clay: alluvium, clay, load, sand and gravel.


3304‐01 / R02 v03 3

Figure 2‐2 Surface Geology Classification (Perth Groundwater Atlas, 2004)

In addition to this regional scale information, a series of 15 soil bores were drilled along the Keane Road alignment by Douglas Partners in November 2008 (full soil bore logs were provided in Appendix D of the KRSL PER document). The bore logs show the presence of fine to medium grained sand along the full length of the alignment, as shown in Figure 2‐3.

The sand layer has a minimum thickness of 0.6 m. Clayey sand underlying the sand layer is present along some sections of the alignment, and generally corresponds to the Swamp & Lacustrine and Guildford Clay geology in Figure 2.2.

The Forrestdale Main Drain Arterial Drainage Strategy (FMDADS) by the Department of Water (2009) states that the superficial geology of the Forrestdale catchment (which includes the Anstley‐Keane Wetlands) consists of degraded, low dunes of Bassendean Sand with low‐lying interdunal areas. The superficial formations generally consist of sandy sediments (a thin layer of Bassendean sand overlying Gnangara sand) with small isolated pockets of clayey sediments (Guildford clay) (Rockwater 2005). The report also notes that the average hydraulic conductivities of the superficial formations have been estimated as ranging from 1.1‐8.9 m/d in the area underlain by Gnangara sand (and Ascot formation), and 0.5‐5.3 m/d in areas of Guildford clay (Rockwater 2005).

Although the soil conditions show the presence of a highly permeable surface sand layer, the modelling assumption adopted in the Water Technology (2013) modelling for the Keane Road project (and in this report) was for no initial loss of rainfall to groundwater. This is conservative, as it assumes a fully saturated groundwater condition (described in the following section) and therefore does not include any infiltration of rainfall into the sand layer.

Swamp & Lacustrine deposits – peat, peaty sand and clay

Guildford Clay: alluvium (clay, loam, sand and gravels)

Bassendean Sand (quartz sand (dunes)

Keane Road


3304‐01 / R02 v03 4

Figure 2‐3 Surface soils stratigraphy and groundwater levels based on borelog data

A site inspection was undertaken to verify the soil type. Sandy soils were noted across the KRSL (Figure 2‐4) alignment as well as surrounding area (Figure 3‐3).

Figure 2‐4 Photo looking along the Keane Road alignment towards Skeet Road


3304‐01 / R02 v03 5

2.3 Groundwater Conditions, Levels and Modelling Assumptions

Groundwater levels are seasonally variable, depending on local rainfall. Groundwater contour information from the Perth Groundwater Atlas is shown in Figure 2‐5. The maximum groundwater elevation and measured groundwater levels are also shown along the Keane Road alignment in Figure 2‐3. The maximum groundwater elevation is similar to the ground surface level, while the limited measured data indicate groundwater levels around 1 m below the ground surface. Due to the thickness of the underlying sand layer and associated high infiltration rates, surface runoff would be expected to occur only under fully saturated groundwater conditions, where the groundwater surface elevation is close to the ground surface.

Figure 2‐5 Groundwater Contours across the Keane Road site (dark blue = maximum groundwater elevation, light blue = groundwater elevation May 2003; Perth Groundwater Atlas http://www.water.wa.gov.au/idelve/gwa/ accessed 04/03/14)

Rockwater Pty Ltd carried out regional groundwater modelling of the Forrestdale main drain catchment in 2005. The report discusses the seasonal variability of groundwater levels in the catchment stating that the average annual groundwater fluctuations in the catchment are generally in the range 1.2 to 2 m, increasing to 2 to 2.5 m near drains and in areas of Guildford clay. Under a high rainfall scenario the winter peak groundwater level in Baileys Wetland was 22.9 m AHD, while the summer minimum was 20.0 m AHD (Department of Water, 2009).

JDA (2012) describes the groundwater system underlying this area for the Heron Park subdivision located to the north of the KRSL as an unconfined aquifer containing generally fresh to slightly brackish groundwater (500 – 1500 mg/L Tota;l Dissolved Solids), with slightly acid to neutral pH (5 – 7). The water table is shallow, rising to within 0.5 m of the surface during winter, depending on surface elevation. The seasonal variation of the water table is about 2m, reaching a maximum elevation in September/October and a minimum elevation in April/May.

These groundwater conditions were accurately incorporated into the Water Technology (2013) modelling as follows:


3304‐01 / R02 v03 6

The application of an approach for assessing the loss of rainfall to groundwater during a storm event (i.e. design losses) which is consistent with previous studies in this area (e.g. JDA, 2002) and the approach recommended in Australian Rainfall and Runoff (1998)

Assumption of no initial losses to simulate an initially wet catchment, which is analogous to winter conditions with a high groundwater elevation.

3. REVISED SURFACE WATER MODELLING

3.1 Purpose

Submissions on the Keane Road Strategic Link PER, have raised concerns that the proposal may not conclusively meet the EPA's objective for hydrological processes as follows:


Specifically, concerns have been raised that the road may interrupt hydrological processes and this could have implications on the ecology of the area. Concerns have also been raised regarding the robustness and verification of hydrological modelling.

As a result of these concerns raised in submissions, further hydrological assessment and model verification has been undertaken in this report, in order to confirm the predicted hydrological impacts and recommended management measures for the project. Further work has included:

Review and confirmation of the appropriateness of previous modelling and management assumptions as described in Section 2 above.

Undertaking a site inspection to verify modelling assumptions and photograph flow path locations.

Re‐running the hydrological modelling using finer scale topography data (LIDAR generated 1 m digital elevation model) provided by Department of Water.

Ground truthing of modelling using aerial photography and rainfall data after a recent series of rainfall events. The high resolution aerial photography clearly shows the spatial extent of surface water inundation within the Keane Anstey Dampland associated with the preceding rainfall events.

3.2 Finer Scale Topography Model Update Overview

This section describes the updated two‐dimension (2D) rainfall on grid modelling of surface water systems across the Anstey‐Keane Wetland for the KRSL, based on the revised 1 m Digital Elevation Model sourced from the Department of Water. This provides significantly higher resolution of the local topography in the hydraulic model compared to the previous investigation (Water Technology, 2013).

The model has been used to simulate the design 100 year Average Recurrence Interval (ARI) rainfall events for the study area, as detailed in the previous report (Water Technology, 2013).

All model parameters from the previous modelling have been adopted for the updated model simulation’s and are described in detail in Water Technology (2013). Modelling assumptions have been justified in Section 2 of this report. It should be noted that the adopted loss values were a proportional loss of 50%, with no initial loss, simulating an initially wet catchment. This is a very conservative assumption, given the presence of sandy soils with high infiltration rates across much of the area and presents the conditions with a fully saturated groundwater system.


3304‐01 / R02 v03 7

3.3 Inflows

Drainage from new residential developments to the north of KRSL mainly infiltrates into soak wells as per the design objectives required by Southern River Integrated Land and Water Management Plan (Department of Water, 2009b) and resulting urban water management planning carried out for these developments. Any overflow drainage during high rainfall events, is directed to the Local Authority drains to the north of KRSL (referred to as the Balannup Drain in the previous report) and by‐passes the catchment upstream of the proposed KRSL.

The only inflows upstream of the proposed KRSL come from the following:

The South East Forrestdale new residential development (Piara Waters), located to the west of the Bush Forever Site 342; and

Baileys Branch Drain, which crosses KRSL near the Anstey Road end.

These inflows and their impacts are discussed in greater detail in the Section 3.3.3.

3.4 Topography

A 1m LiDAR derived Digital Elevation Model (DEM) was sourced from the Department of Water (DOW). This is significantly more detailed that the original model terrain. The topography of the area along with the model extents are shown in Figure 3‐1.


3304‐01 / R02 v03 8

Figure 3‐1 Site Topography


3304‐01 / R02 v03 9

3.5 Flow Paths

3.5.1 Flow Path Delineation

In the original modelling, ESRI’s ArcHydro tool with additional post processing in Matlab was used initially to indicate the likely flow paths across the catchment. This process was repeated using the updated DEM data for the site.

Due to the flat topography and the presence of linear features such as vehicle tracks and disturbances across the site, which interrupt the natural site contours, the flow paths are not always interconnected, as shown in Figure 3‐2.

In some areas, the analysis indicates that flows are directed to local depressions, which when full would overflow towards the drain network across the site.

Figure 3‐3 shows some examples of the existing landscape disturbances which are currently altering the flow paths across the dampland. The flow paths shown are as red arrows in Figure 3‐3 and based on the modelled results described in the following section. Also shown in the figure are the inflows to the Anstey‐Keane Wetland upstream of the KRSL, as noted in Section 3.1 and discussed in detail in Section 3.3.3.

The flow paths identified via the modelled re‐run are consistent with previous modelling results (Water Technology, 2013).

Based on flow paths identified, it is concluded that the dampland is currently highly modified by unsealed roads/tracks which have become un‐natural man made drainage channels (Figure 3‐3). These unsealed roads/tracks currently cause un‐natural surface water blockage, ponding and channelized flow in areas of the dampland which would have previously received natural sheet flow (diffuse flow) prior to human modification via the construction of unsealed tracks and ongoing use of these tracks for off road driving.


3304‐01 / R02 v03 10

Figure 3‐2 Flow Path Delineation


3304‐01 / R02 v03 11

Figure 3‐3 Aerial image of the Keane Anstley Dampland including local site photos


3304‐01 / R02 v03 12

3.5.2 Modelled Flow Paths

The modelled flow paths and flow depth results are shown in Figure 3‐4 and Figure 3‐5 for the critical duration 100 year and 10 year ARI events respectively.

General Flow Path Description

The hydrological regime of a dampland is such that the majority of runoff infiltrates the ground during and/or within a short time after a rainfall event. In some areas water ponds and then gradually infiltrates at a rate dependent on the hydraulic conductivity of the soil, and the underlying groundwater level. For the model simulations it is conservatively assumed the soil is saturated and no infiltration to groundwater occurs over the duration of the event.

In areas where there are no local drains, waterways or unsealed roads/tracks, water flows across the wetland surface in a thin film known as sheet flow. As the model utilizes rain on grid, and has relatively flat topography, sheet flow is widespread and becomes channelized flow in the locations of unsealed roads/tracks. At various localised depressions around the site surface water ponds, reducing the flow velocities. Any surface water depth less than 0.1m is considered sheet flow. On a dry catchment, with the groundwater table at around 1 m below the surface, as indicated from the limited measured data, this sheetflow would infiltrate directly to the groundwater system and no surface flow would be visible.

The Baileys Branch (BB) Drain flows through the Anstey Keane Dampland from just upstream of the Anstey Road end of the proposed KRSL and then downstream of KRSL to Ranford Road (Figure 3‐2). Downstream of Keane Road, the BB Drain passes through a large depression known as Baileys Wetland (Figure 3‐2). Baileys Wetland receives inflows from the urban developments to the north of Bush Forever Site 342 (residential areas of Harrisdale and Piara Waters) via Balannup Drain (which bypasses the KRSL catchment) (Figure 3‐3). Baileys Wetland also receives inflows from the much older residential area of Forrestdale via the BB Drain which overtops when the drain reaches capacity and flows into Baileys Wetland downstream of KRSL. Surface runoff is also directed towards Bailey’s Wetland along existing unsealed roads/tracks in the vicinity of Baileys Wetland (Figure 3‐2).

As noted in the Section 3.5.1 (flow path delineation), across the current site the natural collection and ponding of surface runoff has been altered by the presence of unsealed tracks or vehicle paths, clearance of vegetation associated with the overhead power transmission line and the inputs of stormwater runoff from the surrounding urban developments. Water ponds along the linear features created by the unsealed tracks or roads, and the stormwater outlets from the urban development enter the wetlands at natural low points. As a result these unsealed roads/tracks currently cause un‐natural surface water blockage, ponding and channelized flow in areas of the dampland which would have previously received natural sheet flow (diffuse flow) prior to human modification via the construction of unsealed tracks and ongoing use of these tracks for off road driving.

For all rainfall events ponding of surface water runoff at the Anstey Road end of the study area currently occurs in the following locations (Figures 3‐4, 3‐5 and 3‐6):

Upstream of the proposed KRSL alignment, on cleared private farmland; and

Downstream of the proposed KRSL alignment, associated with the immediately adjacent unsealed tracks.

The extent of ponding in this area is caused by the current capacity of the existing BD Drain itself. The water surface elevation and ponding extent predicted by this report is consistent with the ponding shown in government modelling of the BB Drain (contained within the Figure 10c, of the Forrestdale Main Drain Arterial Drainage Strategy Report, Department of Water, 2009a). It should


3304‐01 / R02 v03 13

be noted that the existing ponding caused by BB Drain is an artificial phenomenon caused by the artificially installed BB Drain. The effect of BB Drain is discussed further in Section 3.3.3.

There are two other locations along the proposed KRSL road alignment where surface flows and ponding currently exist, which are associated with vehicle tracks, collecting and directing flows from upstream. The flow depths and extents in these locations are limited (Figures 3‐4, 3‐5 and 3‐6).

Model Validation against Site Observations and October 2013 Rainfall event

In addition to re‐running the model using finer scale topography (1 m DEM), a further validation step taken as part of this report has been ground truthing of the model as follows:

Conducting a site inspection to confirm the locations of topographic low points and existing artificial disturbances (such as unsealed tracks) causing channelized flow (Figure 3‐3);

Comparing surface water inundation clearly visible in high resolution aerial photography after a recent rainfall event with model outputs. The high resolution aerial photography clearly shows the spatial extent of surface water inundation within the Keane Anstey dampland associated with the preceding rainfall event (Figure 3‐3).

The 10 year ARI map, Figure 3‐6 also includes the aerial imagery from October 2013. There had been a series of rainfall events prior to when the image was captured (refer Figure 3‐7).

The following features are clearly visible in the aerial imagery (Figures 3‐3 and 3‐6):

Surface water ponding and channelised flow along artifical unsealed tracks;

Lower topographic features assocated with lower lying areas (indicated by the presence of greener vegetation).

A GIS based review of the high resolution aerial imagery against the model results has found that the surface water distribution across the study area visible in the image is well represented in the model results.


3304‐01 / R02 v03 14

Figure 3‐4 Maximum water depths for the 100 year ARI rainfall event 9 hour critical duration


3304‐01 / R02 v03 15

Figure 3‐5 Maximum water depths for the 10 year ARI rainfall event 9 hour critical duration


3304‐01 / R02 v03 16

Figure 3‐6 Water depths for the 10 year ARI critical duration event compared to aerial imagery from September 2013


3304‐01 / R02 v03 17

Figure 3‐7 Rainfall Depths at Armadale (9001) for 2013

3.5.3 Impacts of External Inflows

As noted, the only inflows upstream of the KRSL come from the following:

System overflow discharges (after heavy rainfall) from the South East Forrestdale new residential development, approximately 600 m away from the proposed KRSL alignment, located to the west of the Bush Forever Site 342 (refer to inflow points A and B shown on Figure 3‐3); and

Baileys Branch (BB) Drain, which crosses the proposed KRSL alignment near the Anstey Road end.

The inflow locations are shown on Figure 3‐3. These inflows and their impacts on the surface water flows across the system are described below:

South East Forrestdale Residential Area

Only minor inflows come from the “South East Forrestdale” development after high rainfall, into Bush Forever Site 342 upstream of KRSL. Flood storage in swales/basins across the development provides adequate storage to attenuate post‐development flow rates to pre‐development conditions, as detailed in the South East Forrestdale Local Water Management Strategy (LWMS) (VDM Environmental, 2010). The peak inflows to the Anstey‐Keane Wetland occur at two locations (Figure 3‐3). Location A has a peak inflow of 0.2 m3/s for the 10 year ARI event and 0.62 m3/s for the 100 year ARI event, which are less than the pre‐development condition, while Location B has a peak inflow of 0.01m3/s for the 10 year ARI event and 0.02 m3/s for the 100 year ARI event (VDM Environmental, 2010), which are equal to the pre‐development conditions.

Both inflow locations area adjacent to natural basins in the Anstey‐Keane dampland and the inflows effectively pond within the basins until the water infiltrates or evaporates. Given the low peak inflow rates and associated inflow volumes, the presence of sandy soils, the distance from KRSL


3304‐01 / R02 v03 18

(approximately 600 m), and the topography of the wetland at the discharge locations, the inflows are insufficient to generate sheet flows in the direction of the proposed KRSL.

As inflow hydrographs were not available for these inflows locations, no inflows were included in the surface water modelling shown in Figure 3‐4 and Figure 3‐5. However, as a sensitivity test a constant peak inflow of 0.62 m3/s at Location A was applied to the model. This is a very conservative assumption as the constant inflow rate overestimates the inflow volume to the Anstey‐Keane Wetland from the urban development, particularly for long duration rainfall events. These conservative results indicated a flow path could be initiated along the existing track which currently directs flows towards Keane Road and hence this potential flow path has been accommodated through provision of an array of 6 x 300 mm culverts (Location B Figure 3‐9) for the KRSL as discussed in Section 3.5.4 below.

Baileys Branch (BB) Drain

As clearly outlined within the Forrestdale Main Drain Arterial Drainage Strategy (FMDADS) Report (Department of Water, 2009), inflows into Baileys Branch Drain come from the established residential development of Forrestdale, which was developed over 20 years ago. The inflows into BB Drain, from the Forrestdale area and flow rates are well established as outlined with the FMDADS Report (Department of Water, 2009). Peak inflows into the Baileys Branch Drain upstream of Anstey Road are 0.65 m3/s and 0.75m3/s for the 10 year and 100 year ARI events respectively (Department of Water, 2009) and these assumptions have been adopted for this assessment.

As noted in Figure 10c of the FMDADS, under existing conditions flooding occurs from BB Drain in the area north of the intersection of Anstey Road and Keane Road for the 100 year ARI event as the peak water level was predicted to be 22.84 m AHD, and 22.41 m AHD for the 10 year ARI event (Department of Water, 2009). This water surface elevation exceeds the channel bank elevation and therefore flooding occurs in the area north of the intersection of Anstey Road and Keane Road as shown in Figure 10c of the report (Department of Water, 2009) – reproduced as Figure 3‐8 below in this report. The area north of the intersection of Anstey Road and Keane Road is a local low point in the topography (Figure 3‐1) which explains why local surface runoff naturally ponds in this location.

In this report, the modelled maximum 100 year ARI water surface elevation at the intersection of BB drain and the proposed KRSL alignment, is 22.56 m AHD for the 100 year ARI event and 22.48 m AHD for the 10 year ARI event. These modelled surface water elevations are consistent with the modelling within the FMDADS Report (Department of Water, 2009).

Therefore it is concluded that inflows from the BB drain and resulting existing ponding north of the intersection of Anstey Road and Keane Road (Figures 3‐4, 3‐5 and 3‐6) have been accurately modelled within this report.


3304‐01 / R02 v03 19

Figure 3‐8 Baileys Branch Drain Existing System Details (Figure 10c reproduced from FMDADS, Department of Water, 2009)


3304‐01 / R02 v03 20

3.5.4 Recommendations to Maintain Hydrological Connectivity and Prevent Increased Ponding


Provide continued hydrological connectivity for the three flow paths / areas of ponding identified which coincide with the proposed KRSL alignment (A, B, and C as shown in Figure 3‐9).

Ensure that existing ponding currently occurring upstream of the proposed KRSL alignment (Figures 3‐4, 3‐5 and 3‐6) is not increased or decreased;

Where possible, create diffuse flow (sheet flow) downstream of the KRSL alignment, in locations where current channelized flow is occurring un‐naturally down artificial unsealed roads/tracks.

This can be achieved as outlined below.

Overland Flow/Ponding ‐ Locations A, B, and C (Figure 3‐9)

The following management measures are recommended for locations A, B, and C (Figure 3‐9):





The recommended design features a series of 300mm diameter pipe culverts located at each of the flow paths (Table 3‐1). Each array of culverts maintains the hydrologic connectivity and ensures peak 100 year flow rates are maintained without overtopping of the roadway.

Recommendations for locations A and B are similar to those which were previously proposed involving six 300mm diameter pipe culverts installed in an array, to maintain existing hydraulic connectivity and downstream “diffuse” flow in these locations.

Location C on Figure 3‐9 is adjacent to the existing BB Drain which currently passes through 2 x 750mm diameter culverts at Keane Road. These culverts have a capacity of around 2.6m3/s. As detailed in the previous report (Department of Water, 2009) and shown in the current model results, ponding of water does already occur upstream of the proposed KRSL (in private farmland) under existing conditions for the 100 year ARI event. This existing upstream ponding within private farmland is connected to some ponding downstream of the proposed KRSL alignment within native vegetation. Therefore instead of changing the capacity of BB drain (which may change the current hydrology including ponding) it is proposed to provide a series of 300mm diameter pipe culverts underneath KRSL near BB drain, to maintain existing hydrological conditions including ponding on both sides of the KRSL under the 100 year ARI event.


3304‐01 / R02 v03 21

Figure 3‐9 100 year ARI event results showing location of proposed culverts on the KRSL to ensure flow connectivity


3304‐01 / R02 v03 22

Table 3‐1 Culvert Design Options for Flow Paths A, B and C

Flow Path Design Flow for 100 year ARI event (m3/s) No. of 300 mm diameter Pipe Culverts

A 0.5 6

B 0.5 6

C 0.5 6

4. CONCLUSIONS

4.1 Local Hydrology

Key aspects of the local hydrology can be summarised as follows:

The Anstey‐Keane wetland area is classified as a “dampland” wetland system in the Wetland Atlas for the Swan Coastal Plain (Hill et al, 1996). A dampland is defined in the Atlas as a seasonally waterlogged basin, based on the geomorphic classification of wetlands for the Perth‐Bunbury region (Semeniuk & Semeniuk, 1995).

A series of 15 soil bores drilled along the Keane Road alignment by Douglas Partners in November 2008 (full soil bore logs were provided in Appendix D of the Public Environmental Review (PER) document) show the presence of fine to medium grained sand (Bassendean Sand) along the full length of the alignment, with a minimum thickness of 0.6 m. Clayey sand underlying the sand layer is present along some sections of the alignment. Average hydraulic conductivities of these soils have been estimated as ranging from 1.1‐8.9 m/d in the area underlain by sand, and 0.5‐5.3 m/d in areas of Guildford clay (Rockwater 2005).

Regional groundwater modelling of the Forrestdale main drain catchment (Rockwater Pty Ltd, 2005) found that under a high rainfall scenario the winter peak groundwater level in Baileys Wetland was 22.9 m AHD, while the summer minimum was 20.0 m AHD (Department of Water, 2009).

These conditions were accurately incorporated into the surface water modelling assessment detailed in Water Technology (2013).

4.2 Revised Surface Water Modelling

Updated two‐dimensional (2D) rainfall on grid modelling of surface water systems across the Anstey‐Keane Wetland for the KRSL was undertaken, based on the revised 1 m Digital Elevation Model sourced from the Department of Water. This provides significantly higher resolution of the local topography in the hydraulic model compared to the previous investigation (Water Technology, 2013).

The model has been used to simulate the design 100 year Average Recurrence Interval (ARI) rainfall events for the study area, as detailed in the previous report (Water Technology, 2013).

The higher resolution local topography used in the model allowed the improved delineation of flow paths, flood extents and flood depths across the site. In addition, the model results were validated against aerial imagery of the Anstey‐Keane wetland, taken in September 2013 after a series of rainfall events.

The results of the updated assessment have indicated the following:

Based on flow paths identified, it is concluded that the dampland is currently highly modified by unsealed roads/tracks which have become un‐natural man made drainage channels (Figure 3‐3). These unsealed roads/tracks currently cause un‐natural surface water blockage,


3304‐01 / R02 v03 23

ponding and channelized flow in areas of the dampland which would have previously received natural sheet flow (diffuse flow) prior to human modification via the construction of unsealed tracks and ongoing use of these tracks for off road driving.

For all rainfall events ponding of surface runoff occurs north of the intersection of Anstey and Keane Road, which is associated with Baileys Branch Drain. The extent of ponding in this area is dependent on the capacity of the drain itself. The water surface elevation and ponding extent shown is consistent with previous government modelling (Department of Water, 2009).


o Provide continued hydrological connectivity for the three flow paths / areas of ponding identified which coincide with the proposed KRSL alignment (locations A, B, and C as shown in Figure 3‐9).

o Ensure that existing ponding currently occurring upstream of the proposed KRSL alignment (Figures 3‐4, 3‐5 and 3‐6) is not increased or decreased;

o Where possible, create diffuse flow (sheet flow) downstream of the KRSL alignment, in locations where current channelized flow is occurring un‐naturally down artificial unsealed roads/tracks.

To achieve this, the following management measures are recommended for locations A, B, and C (Figure 3‐9):





The recommended design features a series of 300mm diameter pipe culverts located at each of the flow paths / areas of ponding (locations A, B and C). Each array of culverts maintains the hydrologic connectivity and ensures peak 100 year flow rates are maintained without overtopping of the roadway.

The results of this assessment are consistent with the previous hydrological assessment (Water Technology, 2013) as follows:

There are three overland flow paths / areas of ponding (locations A, B and C) across the proposed KRSL alignment.

One artificial channelised flow path also exists across the proposed KRSL alignment – Bailey’s Branch Drain (termed flow path D in the 2013 Water Technology report).

A series of 300m diameter pipe culverts provided at locations A, B, and C will ensure the connectivity of these flow paths / areas of ponding across the proposed KRSL alignment.

The inclusion of an adequate number of appropriately sized culverts at the three locations, as identified in this report, will ensure there are no impacts on the local surface water system.

There are no impacts of the road on the local groundwater or surface water system as the road design allows for the provision of culverts to maintain flow connectivity; the un‐kerbed profile with swale arrangement facilitates infiltration; and the use of the swale for water quality treatment along the alignment will minimise water quality impacts.


3304‐01 / R02 v03 24

As there are no impacts from the road to surface water flow paths and no impacts on groundwater level, flows or water quality, it is unlikely there would be impacts to groundwater dependent or surface water dependent vegetation or ecosystems.

On the basis of the City of Armadale’s commitment to install the recommended management measures (water connectivity culverts) it is concluded the project can meet the EPA objective for Hydrological Processes: To maintain the hydrological regimes of groundwater and surface water so that existing and potential uses, including ecosystem maintenance, are protected.


3304‐01 / R02 v03 25

5. REFERENCES

Davidson, W.A. and Yu, X, 2005, Perth Region Aquifer Modelling System ‐ PRAMS, Hydrogeology and groundwater modelling, Western Australia Department of Environment, Hydrogeology Report No. 202.

Department of Environment and Conservation (2012). ‘Wetland hydrology’, in A guide to managing and restoring wetlands in Western Australia, Prepared by C Mykytiuk, J Lawn and R Lynch, Department of Environment and Conservation, Perth, Western Australia

Department of Water, (2009a). Forrestdale main drain arterial drainage strategy

Department of Water (2009b). Southern River integrated land and water management plan

Hill, Semeniuk, Semniuk and Del Marco, (1996). Wetlands of the Swan Coastal Plain, Volume 2b Wetland mapping, Classification and Evaluation, Wetland Atlas, Water and Rivers Commission

JDA (2002). Southern River/Forrestdale/Brookdale/Wungong Structure Plan, Urban Water Management Strategy, Technical Appendix B Flood Management Modelling

JDA (2012). Heron Park (Phase 2), Harrisdale Local Water Management Strategy (LWMS), Revised to support LSP Amendment, report submitted to Armadale City Council

Rockwater Proprietary Limited (2005) Southern River development area groundwater model, Perth

Semeniuk, V & Seminiuk, CA 1995, 'A geomorphic approach to global classification for inland wetlands. Vegetatio', Vegetatio, vol. 118, no. 118, pp. 103‐124

Smith, A.J .and Pollock, D.W., 2010. Artificial recharge potential of the Perth region superficial aquifer: Lake Preston to Moore River. CSIRO: Water for a Healthy Country National Research Flagship

VDM Environmental, (2010). Urban Water Management Plan Lot 22 Nicholson Road Piara Waters, Report prepared for Mammoth Nominees

keane road strategic link study to public review...keane wetland for the krsl was undertaken, based...

Documents