setting aspirational, resource and management action ... · approach) regression. if...

1

Setting Aspirational, Resource and Management Action Targets Across the Glenelg Hopkins CMA.

4th Milestone Report

January 2005

Prepared by:

Chris Smitt, Jim Cox, Peter Dahlhaus, Darren Bennetts and David Heislers

1

© 2005 CSIRO To the extent permitted by law, all rights are reserved

and no part of this publication covered by copyright may be

reproduced or copied in any form or by any means except with the

written permission of CSIRO Land and Water.

Important Disclaimer

CSIRO Land and Water advises that the information contained in this

publication comprises general statements based on scientific

research. The reader is advised and needs to be aware that such

information may be incomplete or unable to be used in any specific

situation. No reliance or actions must therefore be made on that

information without seeking prior expert professional, scientific and

technical advice.

To the extent permitted by law, CSIRO Land and Water (including its

employees and consultants) excludes all liability to any person for

any consequences, including but not limited to all losses, damages,

costs, expenses and any other compensation, arising directly or

indirectly from using this publication (in part or in whole) and any

information or material contained in it.

Cover Photographs (courtesy of Peter Dahlhaus and Robert

Alexander)

1

INTRODUCTION ....................................................................................................................... 4

BACKGROUND ........................................................................................................................ 5

CONTENT OF PREVIOUS REPORTS ............................................................................. 5 METHODS USED TO SET ASPIRATIONAL, RESOURCE AND MANAGEMENT ACTION

TARGETS ................................................................................................................. 5 STATISTICAL ANALYSIS ............................................................................................. 5 CONCEPTUAL HYDROGEOLOGICAL MODELS ............................................................... 8 GROUNDWATER MODELLING ..................................................................................... 8

PROCESS OF NUMERICAL MODELLING............................................................................ 10

GROUNDWATER MODELLING - SCENARIOS .............................................................. 10 ASSET RISK ASSESSMENT...................................................................................... 12 RESOURCE CONDITION TARGETS ........................................................................... 13

RESULTS OF NUMERICAL MODELLING FOR SUB-CATCHMENTS................................. 15

SUB-CATCHMENT H3.3 (HOPKINS RIVER MUSTONS CREEK) .................................... 15 Description ......................................................................................................................15 Salinity Action Plan Ranking...........................................................................................15 Problem GFS ..................................................................................................................15 Asset Risk Assessment ..................................................................................................16 Salinity processes ...........................................................................................................17 Groundwater modelling simulation and results...............................................................18 Summary.........................................................................................................................21

SUB-CATCHMENT H4.4 (GLENELG RIVER/DUNDAS TABLELANDS)............................. 23 Description ......................................................................................................................23 Salinity Action Plan Ranking...........................................................................................23 Problem GFS ..................................................................................................................23 Asset Risk Assessment ..................................................................................................24 Salinity Processes...........................................................................................................25 Groundwater modelling simulations and results.............................................................25 Summary.........................................................................................................................28

SUB-CATCHMENT G6.1 (GLENELG RIVER/GRAMPIANS HEADWATER) ....................... 29 Description ......................................................................................................................29 Salinity Action Plan Ranking...........................................................................................29 Problem GFS ..................................................................................................................29 Asset Risk Assessment ..................................................................................................30 Salinity Processes...........................................................................................................31 Groundwater modelling simulations and trends .............................................................32 Summary.........................................................................................................................34

SUB CATCHMENT G10.1 (WANNON RIVER – DWYERS CREEK TO FALLS) .................. 35

2

Description ......................................................................................................................35 Salinity Action Plan Ranking...........................................................................................35 Problem GFS ..................................................................................................................35 Asset Risk Assessment ..................................................................................................36 Salinity Processes...........................................................................................................37 Groundwater modelling simulations and trends .............................................................37 Summary.........................................................................................................................39

SUB CATCHMENT G11.2 (WANNON RIVER – GRAMPIANS HEADWATERS).................. 41 Description ......................................................................................................................41 Salinity Action Plan Ranking...........................................................................................41 Problem GFS ..................................................................................................................41 Asset Risk Assessment ..................................................................................................42 Salinity Processes...........................................................................................................43 Groundwater modelling simulations and trends .............................................................43 Summary.........................................................................................................................47 Tube A vs Tube B ...........................................................................................................47

SUB-CATCHMENT H7.3 (MID MT EMU CREEK) ........................................................ 49 Description ......................................................................................................................49 Salinity Action Plan Ranking...........................................................................................49 Problem GFS ..................................................................................................................49 Asset Risk Assessment ..................................................................................................50 Salinity Processes...........................................................................................................51 Groundwater modelling simulations and results.............................................................51 Summary.........................................................................................................................52

SUB-CATCHMENT G11.6 (WANNON RIVER – GRAMPIANS HEADWATERS) ................. 53 Description ......................................................................................................................53 Salinity Action Plan Ranking...........................................................................................53 Problem GFS ..................................................................................................................53 Asset Risk Assessment ..................................................................................................54 Salinity Processes...........................................................................................................55 Groundwater modelling simulations and trends .............................................................55 Summary.........................................................................................................................57

SUB-CATCHMENT G12.1 (BRYANS CREEK) ............................................................. 59 Description ......................................................................................................................59 Salinity Action Plan Ranking...........................................................................................59 Problem GFS ..................................................................................................................59 Asset Risk Assessment ..................................................................................................60 Salinity Processes...........................................................................................................61 Groundwater modelling simulations and trends .............................................................61 Summary.........................................................................................................................63

PROCESS OF SETTING ASPIRATIONAL, RESOURCE AND MANAGEMENT ACTION TARGETS................................................................................................................................ 65

3

SETTING TARGETS ................................................................................................. 65 ASPIRATIONAL TARGETS ........................................................................................ 65 RESOURCE CONDITION TARGETS ........................................................................... 65 MANAGEMENT ACTION TARGETS ............................................................................ 66 STATISTICAL ANALYSIS........................................................................................... 67

Hopkins River at Hopkins Falls.......................................................................................69 Hopkins River at Wickliffe ...............................................................................................71 Glenelg River at Sandford...............................................................................................73 Wannon River at Henty...................................................................................................75

REFERENCES ........................................................................................................................ 77

4

INTRODUCTION This is the fourth and final milestone report on setting Aspirational, Resource and

Management Action Targets across the Glenelg-Hopkins Catchment Management Authority

(GHCMA). The general objective of this project was to assess, spatially re-define and

prioritise the aspirational, resource and management action targets that have been set for the

GHCMA region within their Regional Catchment Strategy.

The specific activity addressed by this milestone report is to present no change conditions

and change conditions under appropriate management for eight base level sub-catchments

(Figure 1), which have been chosen as high risk from 132 across the GHCMA. These base

level sub-catchments will be used as a basis for assessing the effectiveness of salinity

management across other sub-catchments with similar hydrogeological attributes.

Figure 1. Selected base level sub-catchments chosen to assess management strategies to reduce salinity.

5

BACKGROUND Content of previous reports Milestone Report 1 contained background information to the project. Groundwater modelling

and catchment response for four base level sub-catchments in the GHCMA were in Milestone

Report 2. Milestone Report 3 summarised the models and tools used to set the targets.

Methods used to set aspirational, resource and management action targets The methods and tools used to set aspirational, resource and management action targets

across the GHCMA included:

• Statistics (GAM analysis) to determine and predict stream EC trends (i.e. are they rising

or falling),

• Development of conceptual hydrogeological models for cross sections across the base

level sub-catchments where assets are at risk of salinity and estimation of

hydrogeological parameters needed for the groundwater modelling, and

• Groundwater modelling using FLOWTUBE to determine whether land management

changes will protect the assets at risk of salinisation.

Statistical analysis Attempts were made to analyse stream salinity trends at each stream gauging stations within

the GHCMA, however, long term stream salinity monitoring has been infrequent and irregular

across most of the GHCMA, making the establishment of salinity trends difficult in many

cases.

The aim of this type of data analysis is to obtain stream salinity trends that are independent of

fluctuations in flow and season, and hence are indicative of the impacts of saline groundwater

inflows caused by catchment salinisation. A standard approach used in water agencies

throughout the world is the non-parametric Seasonal Kendall’s τ and LOESS smoother

techniques (Hirsch et al., 1982; Cleveland, 1994). As highlighted in Jolly et al. (2001),

problems arise from the application of the Seasonal Kendall’s τ technique when the stream

salinity data are autocorrelated, as was shown to be the case for many of the stations in the

Murray Darling Basin. To overcome this problem a new semi-parametric statistical

methodology was employed (Morton, 1997) which uses the Generalised Additive Model

(GAM) approach (Hastie and Tibshirani, 1990). While corrections for flow and seasonal

effects are implicit in the technique, it is important to note that the analysis does not account

for long-term climatic variations per se.

In this technique, additive regression terms were fitted to logEC (log µS/cm), the explanatory

variables being time (months), logflow (log ML/day) and sinusoidal seasonal terms. This non-

6

linear GAM model represented the response of logEC to time and logflow by arbitrary smooth

curves using cubic splines with knots at each data point. The mathematical form of the

regression was:

( ) ( )( ) ( ) επγπβ

α

+++

++=

122cos122sin

;;

monthmonth

dflogflowSdftimeSlogEC ft (1)

where logEC was the natural logarithm of EC, logflow was the natural logarithm of flow + 1,

time was in years, month had values of 1 to 12, S(t; dft) was a smoothing spline of logEC

versus time with dft degrees of freedom, S(logflow; dff) was a smoothing spline of logEC

versus logflow with dff degrees of freedom; α, β, γ were linear regression coefficients and ε

was the residual error. The terms dft and dff are smoothing parameters that determine the

shape of the splines fitted to the data.

We followed Morton’s (1997) recommendation that values of 4 for dft and 2 for dff were

adequate for data sets of the length used in this project. The term S(x; m) is the sum of the

linear (which is of the form a+b*x) and the non-linear (which has mean zero and no linear

trend) components of the trend. By separating the linear and non-linear components of the

spline function, Equation (1) was rewritten as:

( ) ( ) επγπβχηα

+++++++=

122cos122sin monthmonthClogflowCtimelogEC

logflow

time (2)

where η and χ were linear coefficients of time and logflow respectively, and Ctime and Clogflow

were the non-linear components of S(time;dft) and S(logflow;dff) respectively. The linear

coefficient of time, η, was used to calculate the percentage change in EC per annum using

the formula:

( ){ }1*100 −ηe (3)

Figure 2 shows the two components of the trend. The significance of the non-linear

component cannot be determined and as such is qualitative. The significance of the linear

component of the trend can be determined quantitatively.

7

Sal

inity

Time

non-linear component

linear component +/- 95% confidence interval

Sal

inity

Time

non-linear component

linear component +/- 95% confidence interval

Figure 2. Conceptual graph showing the linear and non-linear components in a flow and seasonally corrected stream salinity trend.

The model fits were carried out using Ordinary Least Squares (referred to as the OLS

approach) regression. If autocorrelation of the residuals of the OLS fits were found to be high

(>0.2), then fits were carried out with first order autoregressive parameters (referred to as the

TSM approach).

The significance of the linear components of the trends at the 5% probability level was

estimated as ±2 standard errors. Standard errors for stations with sufficient data to use the

TSM approach were taken directly from the statistical output. However, if the number of

missing months was too large (>20%), then the TSM approach failed and it was necessary to

use the OLS approach with a multiplier applied to the standard errors. This multiplier was

derived from likelihood theory and was adjusted both for the magnitude of the autocorrelation

and for the amount of missing data:

( ){ } 21

11/21 ARpAR −+ (4)

where p was the proportion of available data and AR1 was the first order autocorrelation

coefficient. It should be noted that this multiplier is an approximation, which assumes that the

missing values occur independently and at random.

For each station, all available EC data were used to determine the trends. Obvious outliers,

caused be erroneous flow or EC values, were identified by high residual values and were

8

removed from the data set and the trends recalculated. There were never more than six

points per station where this was necessary.

Conceptual hydrogeological models A workshop was initially held prior to this project to develop the groundwater flow systems

(GFSs) across the GHCMA (Dahlhaus et al. 2002). A small group met during the course of

this project to then develop appropriate conceptual models of groundwater flow and

hydrological parameters needed for the FLOWTUBE modelling for the base level sub-

catchments chosen. In some instances there was little information to substantiate some of the

conceptual models. In these cases, it was acknowledged that the information was best guess

based on the workshop participants knowledge of the area.

Groundwater modelling Catchment scale groundwater modelling is used to examine the effect of different recharge rates on

the extent of salinity. Groundwater heads along transects across high risk assets1 in base level

sub-catchments in the GHCMA will be simulated using FLOWTUBE (Dawes et al. 1997, Dawes et

al. 2000, Dawes et al. 2001). This is a simple numerical one-dimensional groundwater flow model.

It is a mass-balance model that solves for a change in hydraulic head induced by recharge and

discharge fluxes, and lateral transfers in the direction of flow. The results of FLOWTUBE are

considered to be a hydraulic head transect along an aquifer.

The model considers a one or two-layer system. In the case of a single layer, the aquifer is

assumed to be unconfined and having variable transmissivity dependent on the saturated thickness

of aquifer. In the case of a two-layer system, the lower layer is assumed to contain any lateral

transmission of water while the upper layer contributes storage capacity only. In this case the lower

layer is usually confined or semi-confined, and how this is conceptualised controls the simulated

mechanism for groundwater discharge.

Water sources considered by FLOWTUBE are (i) point sources of runoff at the upstream end of the

aquifer, often manifested as recharge beds collecting surface water from a steeper part of the

catchment, and (ii) diffuse recharge or discharge spread in an arbitrary spatial and temporal pattern

across the aquifer being modelled. The latter source is the recharge component most altered by

the replacement of native species with annual cropping and grazing systems in Australia.

FLOWTUBE allows a variety of boundary conditions for the aquifer. At the downstream end there

are two options: (i) the flux is controlled by a specified groundwater head at a nominated distance,

1 An earlier version of the Geospatial Salinity Hazard and Asset Risk Prediction (GSHARP) process

was used to identify the assets at risk in each sub-catchment (Heislers and Brewin, 2003).

9

useful where a permanent water source occurs nearby that controls head build up such as a river

or irrigation area, or (ii) the flux is controlled by local aquifer properties and the groundwater surface

and drains freely, which can be useful when the groundwater catchment is poorly defined or only

the upper part of a catchment is considered.

For diffuse recharge input there are three options: (i) a user-specified pattern of fixed recharge

amounts with an evaporation component controlled by an extinction depth, which is the traditional

implementation within groundwater models, (ii) recharge calculated by FLOWTUBE internally as

the difference between the current head and a reference elevation multiplied by an impedance

factor, applicable where definite connection and transfers exists between a surface storage and

transmitting aquifer with little outside influence, and (iii) a continuous function of recharge/discharge

based on GIS analysis of depth to water from a DEM, most useful when there are near surface

water levels causing discharge controlled by topographic features.

It should be noted that with the current version of FLOWTUBE, the continuous recharge/discharge

functions are mutually exclusive with the head-induced and fixed recharge pattern, as are the two-

aquifer flux conditions. This means it is not possible to switch between recharge/discharge

functions and a sudden flood or drought through a fixed recharge distribution within the one

simulation. Spikes of recharge however may be superimposed on head-induced recharge

situation.

The aim of the modelling exercise is to improve our understanding of the groundwater processes

and catchment characteristics along with examining land use change with groundwater behaviour

(water balances). The sub-catchments where sufficient data exists will be identified and will be

chosen in such a way that it will represent each GFS within the GHCMA. Each sub-catchment

chosen will have a fairly close series of networked bores with sufficient data to accurately describe

the responsive nature if the GFS to certain land use changes.

Several hypothesis will be tested by the FLOWTUBE modelling:

• to elucidate the hydrogeological processes operating within the sub-catchments and thereby

understand the causes of the salting in affected areas of the GHCMA;

• to quantify the relationship between these hydrogeological processes by simulating them

using a groundwater model of the base level sub-catchments;

• to use this model to test the efficacy of different salinity management strategies.

Sensitivity analysis will also be carried out to demonstrate depth of weathering (aquifer thickness);

width of valley/drainage line and aquifer properties (calibrate using rear heads).

10

FLOWTUBE was attempted in all (8) base level sub-catchments selected (see Table 8) but it was

acknowledged where the model could not be calibrated and where the model would probably not

apply because the hydrogeology is too complex.

PROCESS OF NUMERICAL MODELLING Groundwater Modelling - Scenarios No-change scenarios were predicted on the basis of the available trends, conceptual models,

numerical modelling (FLOWTUBE), or mapped salinity as appropriate for each area. In some

areas the paucity of data resulted in more speculative scenarios. The change scenarios

were based on the adoption of appropriate management actions, rather than the actual likely

adoption rates.

11

Table 1 summarises the base level sub-catchments within the GHCMA chosen for the

modelling component of this study. The assets at risk, a brief summary of why the base level

sub-catchment was chosen and its priority rankings are also shown.

12

Table 1. Sub-catchments in the GHCMA with high asset risk ratings and high groundwater modelling priorities.

Cat ID Assets at

risk Summary Priority

H3.3

Environment

Agricultural

land

Very large and complex catchment, which includes

the well-studied areas of Woorndoo and

Glenthompson. This is a priority area for modelling,

with reference to previous undertaken throughout

the area.

A1

H4.4

Infrastructure

Agricultural

land

Public Land

Medium scale catchment with salinity associated

with the boundary of GFS 13 and GFS 14 and GFS

1. An area where recharge control should be

tested.

A2

G6.1

Environment

Infrastructure

Agricultural

land

Large catchment with significant salinity all in GFS

10 (East Dundas). This includes the Red Barren

salinity field site. High priority for modelling as it

feeds Rocklands Reservoir.

A1

G10.1

Infrastructure

Agricultural

land

Very large catchment with significant areas of

salinity associated with water bodies in GFS 1 and

GFS 10

A1

G11.2

Infrastructure

Agricultural

land

Narrow catchment with significant areas of salinity

associated with the catchment outlet in GFS 14 A2

H7.3

Environment

Infrastructure

Agricultural

land

Large catchment with significant areas of salinity

(mostly primary) associated with wetlands and lakes

in GFS 2

A1

G11.6

Infrastructure

Agricultural

land

Long and narrow with significant areas of salinity

associated with wetlands and lakes in GFS 14 A1

G12.1 Environment,

Agriculture

A large catchment with significant valley floor

salinity mapped in GFS 10, GFS 14 and GFS 17. A2

Asset Risk Assessment Each asset type (agricultural land, environmental, infrastructure) was assessed against the

appropriate hazard criteria to produce a normalised assessment of the risk of salinity to

assets in the sub-catchment (Heislers and Brewin 2003). The conceptual hydrogeological

models were developed along transects which passed through these assets. Flowtubes were

13

chosen so that the management changes which were used in the FLOWTUBE modelling had

most impact on the protection of these assets.

Resource Condition Targets The GHCMA region together with the neighbouring Corangamite CMA has been designated

one of 21 priority regions in Australia in the National Action Plan for Salinity and Water

Quality. The NAP and its Intergovernmental Agreement (IGA) require the GHCMA to comply

with the National Framework for Natural Resource Management (NRM) Standards and

Targets. The framework states a minimum set of regional targets to be developed for:

• Land salinity

• Soil condition

• Native vegetation communities’ integrity

• Inland aquatic ecosystems integrity (rivers and other wetlands)

• Estuarine, coastal and marine habitats integrity

• Nutrients in aquatic environments

• Turbidity / suspended particulate matter in aquatic environments

• Surface water salinity in freshwater aquatic environments

• Significant native species and ecological communities

• Ecologically significant invasive species

A difficulty in setting achievable resource condition targets for some areas is that salinity

trends are not evident in the short monitoring record (or in some cases, no monitoring record).

In other cases, where a trend is measurable, there is no knowledge of what is causing the

trend. In these cases the targets have been set as a commitment to establishing a

measurable target within a reasonably short timeframe.

The timeframe for the achievement of resource condition targets is 10 to 20 years, and they

are intended to be pragmatic and achievable (GHCMA 2002).

Resource Condition Targets also often referred to as End of Valley Targets provide specific,

time bound, measurable targets for the medium term (10-20 years). The National Framework

for natural resource management standards and targets identifies a minimum set of eight

matters for which targets must be set in the region. Three of these relate to salinity

management.

• Area of land threatened by shallow or rising watertables;

• Surface water salinity; and

• Extent of critical assets identified and protected from salinity and degrading water

quality.

14

Establishment of appropriate targets requires prediction of trends, an assessment of risk to

assets and values and agreement on the acceptable level of risk. Insufficient data currently

exists in the region to enable appropriate targets to be set for all matters except surface water

salinity. High priority actions have been identified in the GHCMA Salinity Plan (GHCMA,

2003) to support development of appropriate targets in consultation with the community by

December 2004.

Interim surface water salinity targets have been set for four catchment points by the GHCMA

Salinity Plan, (GHCMA, 2003). Targets indicate the maximum stream salinity level desired for

these rivers by 2012.

• Hopkins River at Wickliffe 15000 EC 90 % of the time;

• Hopkins River at Hopkins Falls 6000 EC 90 % of the time;

• Glenelg River at Sandford 3300 EC 90 % of the time; and

• Wannon River at Henty 5840 EC 90 % of the time.

These targets may change dependant on our understanding of impact of the on-ground works

in altering the resource condition. Information that will enable us make this assessment will

come from modelling land use change and observing the groundwater response (eg the

impact of trees or pastures on salinity). Extrapolation to broader catchment scale, and the

extended timeframe over which actions may become effective or have an impact, add

additional complexity.

15

RESULTS OF NUMERICAL MODELLING FOR SUB-CATCHMENTS Sub-Catchment H3.3 (Hopkins River Mustons Creek) Description Base level sub-catchment H3.3 is very large and complex and includes two well-studied

areas, Woorndoo and Glenthompson (Dixon, 1989, and Cox et al 1999). Because of the

complexity, the results from the modelled area, based on the hydrogeological conceptual

model, cannot be extrapolated to other areas in the sub-catchment. In the Final Report a

number of conceptual models will be developed for the sub-catchment and FLOWTUBE run

for transects within each. This is the only way all assets within the base level sub-catchment

can be assessed as to how they will be affected by land management changes.

Salinity Action Plan Ranking H3.3 base level sub-catchment is Priority A1: salinity hazard, high/moderate value assets,

groundwater system responds to recharge control activities. Options that may work and

require modelling include: Recharge management, discharge management and engineering.

Problem GFS The problem GFS in this base level sub-catchment are (from Dahlhaus et al., 2002):

• GFS 1 – Local Flow Systems in Quaternary Alluvium and coastal deposits;

• GFS 4 – Local Flow Systems in Deeply Weathered Granitic Rocks;

• GFS 5 – Local and Intermediate Deeply Weathered Palaeozoics;

• GFS 8 – Local Flow Systems in the Woorndoo; and

• GFS 14 – Regional and Intermediate Flow Systems in Volcanic Plains Basalt.

16

Figure 3. Location of Sub-Catchment H3.3 and the area targeted for modelling.

Asset Risk Assessment Table 2 to Table 4 identify the salinity hazard rankings, the asset risk ratings and the intended

management options for base level sub-catchment H3.3 (source: Heislers and Brewin, 2003).

Table 2. Salinity hazard rankings for sub-catchment H3.3.

Hazard Rating

Area of discharge Moderate

Line discharge Very Low

Water table < 5 m Moderate

TDS of groundwater Moderate

Flow weighted salinity Very Low

Land management Unit Moderate

Bulk Score 1

17

Table 3. Asset risk ratings for sub-catchment H3.3.

Asset Rating

Road Very High

Rail N/A

Town N/A

Infrastructure Total High

Agricultural Very High

Public Land Low

VROT2 Moderate

Wetlands Moderate

Environment Total High

Average Risk Very High

Water Quality Risk Yes

Table 4. Intended management for sub-catchment H3.3 (source: GHCMA SMP).

Management option (type) Management option (Y/N)

Recharge control Yes

Discharge management Yes

Salinity processes Figure 4 shows a simplified conceptual model of the hydrogeology in the targeted area of sub-

catchment H3.3. This conceptual model has been based on a detailed study in the Woorndoo

area, Stuart-Smith and Black (1999) and Paine et al. (2004), with further input from key

professionals during a workshop held in Ballarat during March.

2 * The actual VROT species threatened is not indicated in Heislers and Brewin (2003)

18

Figure 4. Conceptual model of the salinity processes within sub-catchment H3.3.

Groundwater modelling simulation and results Three basic scenarios have been simulated for the base level sub-catchment:

• Reduce recharge rates by 90% in the top quarter of the catchment;

• Reduce recharge rates by 90% in the middle third of the catchment; and

• Reduce recharge rates by 90% over the entire catchment.

Figure 5 shows the simulated location of the water levels for continual recharge that has been

reduced by 90% in the top quarter of the catchment. It should be noted that after 100 years

there is no change in the water level, suggesting that the system has reached a new

equilibrium.

19

160

170

180

190

200

210

220

230

240

250

260

0 500 1000 1500 2000 2500 3000 3500 4000 4500Distance across catchment (m)

Elev

atio

n (m

, AH

D)

TopographyWater TableBasementafter 5 yrsafter 10 yrsafter 20 yrsafter 50 yrsafter 100 yrs

Groundwaterdischarge to lakes

Low Recharge

Trees planted1 mm/yr recharge

Figure 5. The effect of reducing the recharge by 90% over the top quarter of the modelled area [water levels are simulated for 5, 10, 20 50 and 100 years into the future].

Figure 6 shows the simulated location of the water levels for continual recharge that has been

reduced by 90% in the middle third of the catchment. In this scenario, no change is seen with

water level reduction after 10 years.

160

170

180

190

200

210

220

230

240

250

260


Elev

atio

n (m

, AH

D)

TopographyWater TableBasementafter 5 yrsafter 10 yrsafter 20 yrs

Groundwaterdischarge to lakes


High Recharge

20

Figure 6. The effect of reducing the recharge by 90% over the middle third of the modelled area [water levels are simulated for 5, 10 and 20 years into the future].

This simulation clearly shows that reducing land use in this area via land use change has very

little impact on future water levels throughout the catchment. This suggests that the majority

of the water that discharges in the lower parts of the catchment is sourced from either the

highland reaches or intermediate through flow from neighbouring catchments (namely base

level sub-catchments H3.4, H4.4 and H11.1).

The final simulation in this sub-catchment is presented in Figure 7. Here the location of the

water levels for continual recharge that has been reduced by 90% over the entire catchment

is shown. In this scenario, there is no change in the water level (i.e. a new equilibrium

reached) after 150 years. This is indicative of an Intermediate Groundwater Flow System

rather than a Local Flow System. Also it should be noted that this level of land use change

significantly lowers through the catchment and after 20 years, discharge to the Woorndoo

Lakes ceases, causing them to potentially dry up, given no direct surface runoff.

160

170

180

190

200

210

220

230

240

250

260


Elev

atio

n (m

, AH

D)

TopographyWater TableBasementafter 5 yrsafter 10 yrsafter 20 yrsafter 50 yrsafter 100 yrsafter 150 yrs

Figure 7. The effect of reducing the recharge by 90% over the entire modelled area [water levels are simulated for 5, 10, 20 50, 100 and 150 years into the future].

Table 5 summarises the net effect between recharge reduction and the associated proportion

of surface discharge for modelled scenarios 1 and 3.

21

Table 5. Modelled results showing reduction of saline area within a FLOWTUBE.

Reduction of area

Recharge to 1mm/yr in top quarter

Reduction of area

Recharge to 1mm/yr entire area Modelled results

(H3.3) (km2) (%) (km2) (%)

Initial state 6 0 6 0

After 5 yrs 4.8 20 3.6 40

After 10 yrs 3.6 40 3 50

After 20 yrs 2.4 60 1.8 70

After 50 yrs 1.8 70 0 100

After 100 yrs 1.8 70 0 100

Summary Within this base level sub-catchment the assets most at risk are the roads, agricultural land

and public land, therefore the modelling targeted the protection of these assets. It is evident

from the FLOWTUBE modelling that recharge reduction significantly impacts water table

elevation throughout the catchment and especially in the upland parts, and depending on the

extent of the land management new equilibriums can be seen from between 50 to 150 years.

This has significant implications on protecting assets such as infrastructure and agricultural

land as the main issues here are shallow saline groundwaters. What is also evident is that if

you reduce recharge in the top quarter of the catchment, you are not only reducing the water

levels down gradient but you are effectively cutting off the supply of fresh water which mixes

with the underlying saline groundwater that discharges into the lakes. Potentially, this means

that if you were to undertake this management option not only would you decrease water

levels in the upland reaches of the catchment but you may also cause the lakes where the

groundwater discharges to become saltier. To overcome this problem engineering drains

may be installed in to rapidly direct the water off the landscape and into these discharge

lakes.

However, within this region there may exist a Victorian rare or threatened species (VROT)

that require no significant changes to the hydrology of the landscape to keep their habitat. In

such a scenario, reducing the water table elevations may cause a change in their habitat so it

is necessary to keep this in mind when undertaking any management actions. Figure 8

shows graphically the relationship between recharge reduction and surface discharge.

22

Figure 8. Area targeted for salinity management and summary of surface discharge response to a 90% recharge reduction across the modelled area.

23

Sub-Catchment H4.4 (Glenelg River/Dundas Tablelands) Description Base level sub-catchment H4.4 is a medium to large catchment with salinity associated with

the boundary of GFS 13 and GFS 14 and GFS 1. This base level sub-catchment is

considered a high priority for modelling as there are several important assets at risk including,

infrastructure, public and agricultural land. Despite this being a priority A2 base level sub-

catchment (i.e. GFS not expected to respond to recharge management), recharge control

should be tested on local areas where assets are at high risk.


groundwater system does not responds to recharge control activities. Options that may work

and require modelling include: Discharge management and engineering.

Problem GFS The problem GFS in this base level sub-catchment is:

• GFS 1 - Local flow systems in Quaternary alluvium and coastal deposits;

• GFS 13 – Intermediate and Local Flow Systems in the Fractured Palaeozoics; and

• GFS 14 - Regional and intermediate flow systems in Volcanic Plains basalt.

Figure 9 shows the location of these GFSs within the base level sub-catchment.

Figure 9. Sub-catchment H4.4 and the area targeted for modelling.

24

Asset Risk Assessment Table 6 to Table 8 identify the salinity hazard rankings, the asset risk ratings and the intended

management options for sub-catchment H4.4 (source, Heislers and Brewin, 2003).


Hazard Rating

Area of discharge Low

Line discharge Very Low

Water table < 5 m High

TDS of groundwater Medium

Flow weighted salinity Low

Land management Unit Medium

Bulk Score 1


Asset at risk Rating

Road Very High

Rail Low

Town Low

Infrastructure Total Very High

Agricultural High

Public Land Medium

VROT3 Low

Wetlands Medium

Environment Total High



3 * The actual VROT spiecies threatened is not indicated in Heislers and Brewin (2003)

25



Recharge control No


Salinity Processes Figure 10 shows the area, which was chosen from within the base level sub-catchment H4.4

to develop a conceptual hydrogeological model and determine how it responds to recharge

management.

Figure 10. Conceptual model of the salinity processes within sub-catchment H4.4

This conceptual model has been developed by key professionals during a workshop held in

Ballarat during March and from limited geological logs throughout the sub-catchment.

Groundwater modelling simulations and results Two basic scenarios have been simulated for the base level sub-catchment:

• Reduce recharge rates by 90% in the top quarter of the catchment; and


Figure 11 shows the simulated location of the water levels for continual recharge that has

been reduced by 90% in the top quarter of the catchment. It should be noted that after 100

26

years there is no change in the water level, suggesting that the system has reached a new

equilibrium.

200

220

240

260

280

300

320

340

360

380

400

0 2000 4000 6000 8000 10000 12000Distance across catchment (m)

Elev

atio

n (m

, AH

D)


Figure 11. The effect of reducing the recharge by 90% over the top quarter of the modelled area [water levels are simulated for 5, 10, 20 50 and 100 years into the future].

This simulation clearly shows that reducing recharge in this area via land use change has

quite a significant impact on highland groundwater levels with the upper half of the catchment

drying right out. However, there is low confidence on aquifer thickness within this base level

sub-catchment as there were only 2 bore logs that contained aquifer depth data. Therefore,

the depth to basement was kept constant at an elevation of 200 m.

The final simulation in this base level sub-catchment is presented in Figure 12. Here the

location of the water levels for continual recharge that has been reduced by 90% over the

entire catchment is shown.

27

200

220

240

260

280

300

320

340

360

380

400

0 2000 4000 6000 8000 10000 12000Distance across catchment (m)

Elev

atio

n (m

, AH

D)


Figure 12. The effect of reducing the recharge by 90% over the entire modelled area [water levels are simulated for 5, 10 and 20 years into the future].

In this scenario, there is no change in the water level (i.e. a new equilibrium reached) after

150 years. This simulation clearly shows that reducing recharge in this area via land use

change has quite a significant impact on highland groundwater levels. However, there is very

little difference between this scenario and that demonstrated in Figure 11. It should be note

that the responsiveness of the aquifer is indicative of a Local Flow System.

Table 9 summarises the net effect between recharge reduction and the associated proportion

of surface discharge for the two modelled scenarios within the modelled area.


Reduction of area


Reduction of area

Recharge to 1mm/yr over entire

area Modelled results

(H4.4)

(km2) (%) (km2) (%)

Initial state 8.40 0 8.40 0

After 5 yrs 8.40 0 6.55 22

After 10 yrs 8.40 0 5.60 33

After 20 yrs 6.55 22 5.60 33

After 50 yrs 5.60 33 5.60 33

After 100 yrs 5.60 33 5.60 33

28

Summary Within this base level sub-catchment the assets most at risk are infrastructure, (roads and

rail), agricultural and public land, therefore the modelling targeted the protection of these

assets. It is evident from the FLOWTUBE modelling that recharge reduction significantly

impacts water table elevation along the steeper slopes of the transect and depending on the

extent of the land management new equilibriums can be seen from between 100 to 150

years. In the flatter part of the transect, recharge reduction seems to have very little effect of

groundwater elevations. Figure 12 shows graphically the relationship between recharge

reduction and surface discharge.

This has significant implications on protecting assets such as infrastructure and agricultural

land in this area as the primary reason they are at risk is due to the presence of the shallow

saline groundwater. This is a region where discharge management becomes a priority and

should be investigated further.


29

Sub-Catchment G6.1 (Glenelg River/Grampians Headwater)

Description Base level sub-catchment G6.1 is a very large and complex catchment that includes the

documented studied areas by Brouwer and Fitzpatrick in 2002. This base level sub-catchment

is considered a high priority for modelling as there are several important assets at risk

including, infrastructure, public land, agriculture and environmental sites.

Salinity Action Plan Ranking G6.1 base level sub-catchment is Priority A1: salinity hazard, high/moderate value assets,




GFS 10 – Local flow systems in the East Dundas.

Figure 14. Location of this GFS within the base level sub-catchment.

30

Asset Risk Assessment Table 10 to Table 12 identify the salinity hazard rankings, the asset risk ratings and the

intended management options for sub-catchment G6.1 (source: Heislers and Brewin, 2003).

Table 10. Salinity hazard rankings for sub-catchment G6.1.

Hazard Rating

Area of discharge Moderate

Line discharge Moderate


TDS of groundwater Moderate

Flow weighted salinity High

Land management Unit Very High

Bulk Score 3

Table 11. Asset risk ratings for sub-catchment G6.1.


Road Very High

Rail N/A

Town Low



Public Land Very High

VROT4 High

Wetlands N/A

Environment Total Very High



Table 12. Intended management for sub-catchment G6.1 (source: GHCMA SMP).




4 * The actual VROT spiecies threatened is not indicated in Heislers and Brewin (2003)

31

Salinity Processes Figure 15 and Figure 16 show conceptual models of the hydrogeology and salinity processes

sub-catchment G6.1. These conceptual models have been based on a detailed study, which

encompasses the study area (Brouwer and Fitzpatrick, 2002a, 2002b).

Figure 15. Conceptual model of salinity process within sub-catchment G6.1.

Figure 16. Conceptual model of salinity process within sub-catchment G6.1.

32

Groundwater modelling simulations and trends Two basic scenarios have been simulated for the base level sub-catchment:

• Reduce recharge rates by 90% in the top quarter of the modelled area; and

• Reduce recharge rates by 90% over the entire modelled area.


been reduced by 90% in the top quarter of the modelled area. It should be noted that after

100 years there is no change in the water level, suggesting that the system has reached a

new equilibrium.

180

190

200

210

220

230

240

250

260

270

280


Elev

atio

n (m

, AH

D)



Figure 17. Shows the effect of reducing the recharge by 90% over the top quarter of the catchment [water levels are simulated for 5, 10, 20 50 and 100 years into the future].

This simulation clearly shows that reducing recharge via land use in this area significantly

impacts on future water levels in the upper two thirds of the area. This suggests that the

majority of the water that discharges in the lower parts of the catchment is source from either

the highland reaches or intermediate through-flow from neighbouring catchments (namely

base level sub-catchments G5.4, G10.2 and/or G6.3). However, at the lower end the model

was forced to merge into a constant head boundary and thus may not be entirely accurate at

this end.

The final simulation in this sub-catchment is presented in Figure 18. Here the location of the

water levels for continual recharge that has been reduced by 90% over the entire modelled

33

area are shown. In this scenario, there is no change in the water level (i.e. a new equilibrium

reached) after 100 years. This is indicative of either an Intermediate Groundwater Flow

System or a Local Flow System. Also it should be noted that this level of land use change

significantly lowers the water table through the catchment. The responsiveness of the aquifer

is quite rapid with rapid falls occurring in the first 10 to 20 years.

180

190

200

210

220

230

240

250

260

270

280


Elev

atio

n (m

, AH

D)


Figure 18. Shows the effect of reducing the recharge by 90% over the entire catchment [water levels are simulated for 5, 10, 20 50 and 100 years into the future].

Table 13 summarises the net effect between recharge reduction and the associated

proportion of surface discharge for the two modelled scenarios within the modelled section.

Table 13. Summary of modelled results showing reduction of saline area within a FLOWTUBE.

Reduction of area


Reduction of area

Recharge to 1mm/yr entire

modelled area Modelled results

(G6.1)

(km2) (%) (km2) (%)


After 5 yrs 4.5 10 1.5 70

After 10 yrs 4 20 1.5 70

After 20 yrs 3 40 1.5 70

After 50 yrs 1.5 70 0.5 90

After 100 yrs 1.5 70 0 100

34

Summary This model behaves similarly to that of H3.3 with the exception being a quicker response to

land use change. Within this base level sub-catchment the assets most at risk are the roads,

agricultural land and public land, therefore the modelling targeted the protection of these

assets. It is evident from the FLOWTUBE modelling that recharge reduction rapidly and

significantly impacts water table elevation throughout the catchment and especially in the

upland parts, and depending on the extent of the land management new equilibriums can be

seen from between 50 to 100 years. Figure 19 shows graphically the relationship between

recharge reduction and surface discharge.

By reducing recharge in the upland reaches of the catchment, assets such as infrastructure

and agricultural land, which are at risk by shallow saline groundwater, may be preserved.

What is also evident (and as in the case of base level sub-catchment H3.3) is that if you

reduce recharge in the top quarter of the catchment, you are not only reducing the water

levels down gradient but you are effectively cutting off the supply of fresh water which mixes

with the underlying saline groundwater that discharges into the Glenelg River. To overcome

this problem engineering drains may be installed in to rapidly direct the water off the

landscape and into the Glenelg River, which would not only “freshen” up the immediate

environment but would also help decrease stream EC levels down stream.

Figure 19. Area targeted for salinity management and summary of surface discharge response to a

90% recharge reduction across the modelled area.

35

Sub Catchment G10.1 (Wannon River – Dwyers Creek to Falls) Description Sub-catchment 10.1 is a large area (approx 390 km2) generally bisected by the Wannon River

south west of The Grampians. The western portion of the catchment comprises the south-

eastern Dundas Tableland (GFS 10), on which the salinity processes have been extensively

studied by the University of Ballarat (Dahlhaus et al., 2000). The salinity in eastern portion of

the sub-catchment occurs within the volcanic plains (GFS 14). The Bulart area was chosen

for FLOWTUBE modelling as typical of the salinity processes for the western portion of the

catchment (GFS10). The conceptual model is based on that published by Dahlhaus and

MacEwan (1997), which comprises a slow moving regional groundwater flow system in the

fractured Rocklands Volcanics overlain by a local system in the deeply weathered saprolite





• GFS 1 – Local Flow Systems in Quaternary Alluvium and coastal deposits;

• GFS 10 – Local Flow Systems in the East Dundas; and


Figure 20. Location of sub-catchment G10.1 and the area targeted for modelling.

36




Hazard Rating

Area of discharge Medium

Line discharge Very High

Water table < 5 m Medium


Flow weighted salinity Medium


Bulk Score 2



Road Very High

Rail N/A

Town N/A



Public Land Medium

VROT5 Medium

Wetlands N/A

Environment Total Medium


Water Quality Risk N/A






37

Salinity Processes Figure 21 shows the area, which was chosen from within the base level sub-catchment G10.1


management.

Figure 21. Conceptual model of the salinity processes within sub-catchment G10.1.





been reduced by 90% in the top quarter of the modelled area. Figure 23 shows the simulated


area.


proportion of surface discharge for the two modelled scenarios within the modelled area.

38

150

170

190

210

230

250


Elev

atio

n (m

, AH

D)


Figure 22. Effect of reducing the recharge by 90% over the top quarter of the catchment [water levels are simulated for 5, 10, 20, 50 and 100 years into the future].

150

170

190

210

230

250


Elev

atio

n (m

, AH

D)


Figure 23. Effect of reducing the recharge by 90% over the entire catchment [water levels are simulated for 5, 10, 20 50 and 100 years into the future].

39


Reduction of area


Reduction of area

Recharge to 1mm/yr catchment

wide Modelled results

(G10.1)

(km2) (%) (km2) (%)


After 5 yrs 0.84 0 0.84 0

After 10 yrs 0.84 0 0.72 14

After 20 yrs 0.84 0 0.6 29

After 50 yrs 0.84 0 0.36 57

After 100 yrs 0.84 0 0.12 85

Summary Within this base level sub-catchment the assets most at risk are the roads and agricultural

land therefore the modelling targeted the protection of these assets. Within the modelled

area there only existed agricultural land as insufficient data was available to model an area

which contained both assets at risk. The two modelled scenarios showed that the water table

fell in the upper parts of the modelled area before reaching a new equilibrium. However, at

the valley end or “outflow” area, the water table remains at a fairly constant level.

The results also suggest that there is a significant advantage in decreasing recharge across

the entire modelled area as opposed to just the top quarter as the reduction of area salinised

or waterlogged decreases from 0% to 85%, however it must be noted that the initial area

waterlogged or salinised is quite small. Figure 24 below shows graphically the relationship

between recharge reduction and surface discharge.

40


41

Sub Catchment G11.2 (Wannon River – Grampians Headwaters) Description Sub-catchment G11.2 (approximately 144 km2) forms the elongated east-west catchment of

Back Creek, south of Dunkeld. This entire sub-catchment lies on the volcanic plain (GFS 14),

which is overlain by a thin veneer of alluvial (stream) and paludal (swamp) sediments in

places (GFS 1). The majority of the salinity has been mapped in the western portion of the

sub-catchment, where Back Creek formerly drained into a series of swamps and ephemeral

wetlands. The swamps and wetlands have now been drained by diverting Back Creek into

the Wannon River to the north. It is likely some of the salinity in the shallow drainage

depressions would be primary.






Figure 25. Location of Sub-Catchment G11.2 and the area targeted for modelling.

42




Hazard Rating


Line discharge Low

Water table < 5 m Low


Flow weighted salinity Medium

Land management Unit Medium

Bulk Score 1



Road Very high

Rail Medium

Town N/A



Public Land Low

VROT6 Low

Wetlands N/A

Environment Total Low





Recharge control No


43




management.


Groundwater modelling simulations and trends Two dominant groundwater flow direction existed in this sub-catchment, and have been

represented by the FLOWTUBES drawn in Figure 25 and Figure 31 and basically follow the

local topography. The modelled results hope to differentiate between which is the primary

flow direction responsible for salinity within this area.

Two basic scenarios have been simulated for the base level sub-catchment, for each tube:



44

Tube A




area.



Tube B




area.



150

175

200

225

250

275

300

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000Distance across catchment (m)

Elev

atio

n (m

, AH

D)


Figure 27. Effect of reducing the recharge by 90% over the top quarter of the catchment [water levels are simulated for 5, 10, 20 50 and 100 years into the future].

45

150

175

200

225

250

275

300

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000Distance across catchment (m)

Elev

atio

n (m

, AH

D)


Figure 28. Effect of reducing the recharge by 90% over the entire catchment [water levels are simulated for 5, 10, 20 50, 100 and 150 years into the future].


Reduction of area


Reduction of area



(G11.2 Tube A)

(km2) (%) (km2) (%)


After 5 yrs 7 6.7 4.5 40

After 10 yrs 6 20 2.5 66.66

After 20 yrs 5 33.33 1.5 80

After 50 yrs 4 46.66 0.5 93.33

After 100 yrs 4 46.66 0.5 93.33

46

150

175

200

225

250

275

300

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000Distance across catchment (m)

Elev

atio

n (m

, AH

D)



150

175

200

225

250

275

300

0 500 1000 1500 2000 2500 3000 3500 4000 4500 5000Distance across catchment (m)

Elev

atio

n (m

, AH

D)


Figure 30. Effect of reducing the recharge by 90% over the entire catchment [water levels are simulated for 5, 10, 20 50, 100 and 150 years into the future].

47

Table 22 Modelled results showing reduction of saline area within a FLOWTUBE.

Reduction of area


Reduction of area

Recharge to 1mm/yr modelled

area Modelled results

(G11.2 Tube B)

(km2) (%) (km2) (%)


After 5 yrs 3.75 50 0 100

After 10 yrs 0.5 93.33 0 100

After 20 yrs 0 100 0 100

After 50 yrs 0 100 0 100

After 100 yrs 0 100 0 100

Summary Within this base level sub-catchment the assets most at risk are the roads and agricultural

land therefore the modelling targeted the protection of these assets. Within the modelled

area there only existed agricultural land as insufficient data was available to model an area

which contained both assets at risk. The two modelled scenarios in both tubes A and B,

showed that the water table fell in the upper parts of the modelled area before reaching a new

equilibrium.

Tube A vs Tube B Tube A and B respond quite differently to the same change in land-use. Initially, both tubes

had a similar proportion of the modelled area salinised or waterlogged, however tube B

responds far more quickly to a particular land-use change than tube A. In tube B, surface

discharge seems to have stopped after just 5 years where as tube A still has ~60%

discharging to the surface when a 90% reduction in recharge is applied.

The results also suggest that there is a significant advantage in decreasing recharge across

areas which have a south to north flowing groundwater flow system. Figure 31 below shows

graphically the relationship between recharge reduction and surface discharge.

48


49

Sub-Catchment H7.3 (Mid Mt Emu Creek) Description Base level sub-catchment H7.3 is a large catchment with significant areas of salinity (mostly

primary) associated with wetlands and lakes in GFS 2. This base level sub-catchment is

considered a high priority for modelling as there are several important assets at risk including,

infrastructure, public land, agriculture and environmental sites.





GFS 2 – Local and Intermediate Flow Systems in the Volcanic plains (later phase volcanics).

Figure 32. Location of this GFS within the base level sub-catchment..

50


intended management options for sub-catchment H7.3, (source: Heislers and Brewin, 2003).


Hazard Rating

Area of discharge High

Line discharge Very High

Water table < 5 m Low


Flow weighted salinity High

Land management Unit High

Bulk Score 3



Road Medium

Rail N/A

Town N/A

Infrastructure Total Low


Public Land High

VROT7 Very High

Wetlands N/A






Recharge control N/A



51

Salinity Processes Figure 33 shows a conceptual model of the hydrogeology for sub-catchment H7.3. This

conceptual model has been developed by key professionals during a workshop held in

Ballarat during March and from limited geological logs throughout the sub-catchment.

Figure 33. Conceptual model of salinity process within sub-catchment H7.3.

Groundwater modelling simulations and results FLOWTUBE modelling within this base level sub-catchment was unsuccessful due to the

complex hydrogeology. FLOWTUBE could not calibrate against the observed water levels, as

the interaction between GFS 2 and 14 seems to be too great. The role the volcanic vent plays

also adds further complexity to the system. In essence, too much water was coming in, either

via recharge in GFS 2 or through-flow from GFS 14. For the model to calibrate against the

observed water levels, unrealistic hydraulic conductivity values of greater than 50 m/d were

required and/or increasing the depth to bedrock to greater than 100m. Also the amount of

water entering the top of the catchment had to be reduced to virtually 0 m/d, which is not

unrealistic for upland local flow systems but in this case there is significant through-flow from

neighbouring catchments.

52

Summary A more complex groundwater flow model such as MODFLOW along with higher array of input

data is required to accurately model future groundwater response to land use change in this

area. However, based on the conceptual model it is expected that by reducing recharge

within GFS 2, you will significantly lower the water table in the upland reaches with the danger

of drying and\or salting up Lake Gellie.

53

Sub-Catchment G11.6 (Wannon River – Grampians Headwaters) Description Approximately 118 km2, sub-catchment G11.6 forms an elongated drainage basin with poorly

developed internal drainage. The boundary encloses the sedimentary rocks (GFS 5) of the

Glenthompson hills in the south and extends almost to Lake Muirhead in the north. The

majority of the catchment lies on the volcanic plains (GFS 14) where almost all of the salinity

is associated with lakes, swamps and broad drainage depressions. Many of the lakes have

accumulated a thin veneer of alluvial (stream) and paludal (swamp) sediments (GFS 1). It is

likely that much of the salinity is naturally occurring (primary).





GFS 1 – Local Flow Systems in Quaternary Alluvium and costal deposits; and

GFS 14 – Regional and Intermediate Flow Systems in Volcanic Plains Basalt.


54




Hazard Rating

Area of discharge High

Line discharge Very low

Water table < 5 m Very High

TDS of groundwater High

Flow weighted salinity Low

Land management Unit Low

Bulk Score 3



Road Very High

Rail N/A

Town N/A



Public Land High

VROT8 Very High

Wetlands N/A




Table 28. Intended management for sub-catchment G11.6 (source, GHCMA SMP).


Recharge control No


55




management.

Figure 35. Conceptual model of the salinity processes within sub-catchment G11.6

Groundwater modelling simulations and trends Three basic scenarios have been simulated for the base level sub-catchment:

• Reduce recharge rates by 90% in the top quarter of the modelled area;

• Reduce recharge rates by 90% in the bottom third of the modelled area; and




water levels for continual recharge that has been reduced by 90% over the bottom third of the

modelled area and Figure 38 shows the simulated location of the water levels for continual

recharge that has been reduced by 90% over the entire modelled area.



56

150

175

200

225

250

275

300

0 1000 2000 3000 4000 5000 6000 7000 8000Distance across catchment (m)

Elev

atio

n (m

, AH

D)



150

175

200

225

250

275

300


Elev

atio

n (m

, AH

D)


Figure 37. Effect of reducing the recharge by 90% over the bottom third of the modelled area [water levels are simulated for 5, 10, 20, 50 and 100 years into the future].

57

150

175

200

225

250

275

300


Elev

atio

n (m

, AH

D)


Figure 38 Effect of reducing the recharge by 90% over the modelled area [water levels are simulated for 5, 10, 20, 50 and 100 years into the future].


Reduction of area


Reduction of area



(G11.6)

(km2) (%) (km2) (%)


After 5 yrs 4.5 10 3.5 30

After 10 yrs 4 20 3.5 30

After 20 yrs 4 20 3.5 30

After 50 yrs 4 20 3.5 30

After 100 yrs 4 20 3.5 30

Summary Within this base level sub-catchment the assets most at risk are environmental assets

including Victorian Rare of Threatened Species (VROT), roads and agricultural land therefore

the modelling targeted the protection of these assets. It is evident from the FLOWTUBE

modelling that recharge reduction has a significant impact on water table elevation throughout

the modelled area except in the vicinity of the lake in the middle if the modelled area. This

would imply that salinity associated with this lake is “primary”. This has significant

implications on protecting assets such as infrastructure and agricultural land as within this

58

region engineering options would be required to lower the shallow saline groundwaters.

However, within this region there may exist a Victorian rare or threatened species (VROT)

that require no significant changes to the hydrology of the landscape to keep their habitat. In

such a scenario, reducing the water table elevations in the vicinity of the lake may cause a

change in their habitat so it is necessary to keep this in mind when undertaking any

management actions.

Figure 39 below shows graphically the relationship between recharge reduction and surface

discharge.


59

Sub-Catchment G12.1 (Bryans Creek) Description The majority of sub-catchment G12.1 (approximately 257 km2) lies on the Western Dundas

Tableland (GFS 9). The western portion of the sub-catchment is drained by Bryan Creek, the

headwaters of which (outside this sub-catchment) are fed by primary saline discharge

(Nathan, 1999). Just west of Coleraine, Bryan Creek is joined by the Konong Wootong

Creek, which drains the central portion of the sub-catchment, before it continues south-west

across the Merino Tableland (GFS 12) to join the Wannon River. The majority of salinity in

this sub-catchment has been mapped in the north-west corner, within the catchment of the

Konongwootong Reservoir. Although the water quality in the reservoir has been adversely

affected by salinity, comparatively little is known about the groundwater flow system and

salinity processes.





• GFS 1 – Local Flow Systems in Quaternary Alluvium and costal deposits;

• GFS 9 – Local Flow Systems in the Worndoo Complex.


60


intended management options for sub-catchment G12.1 (source, Heislers and Brewin, 2003).


Hazard Rating


Line discharge High


TDS of groundwater High

Flow weighted salinity Very High


Bulk Score 5



Road Very High

Rail N/A

Town Low


Agricultural High

Public Land Very High

VROT9 Very High

Wetlands N/A









61



management.



• Reduce recharge rates by 90% in the top quarter of the catchment; and




location of the water levels for continual recharge that has been reduced by 90% over the

entire modelled area.

Table 33 Summarises the net effect between recharge reduction and the associated


62

200

225

250

275

300

325

350

0 400 800 1200 1600 2000Distance across catchment (m)

Elev

atio

n (m

, AH

D)


Figure 42. Effect of reducing the recharge by 90% over the top quarter of the modelled area [water levels are simulated for 5, 10, 20, 50 and 100 years into the future].

200

225

250

275

300

325

350

0 400 800 1200 1600 2000Distance across catchment (m)

Elev

atio

n (m

, AH

D)


Figure 43. Effect of reducing the recharge by 90% over the entire modelled area [water levels are simulated for 5, 10, 20, 50 and 100 years into the future].

63


Reduction of area


Reduction of area



(G12.1)

(km2) (%) (km2) (%)


After 5 yrs 0.675 10 0.3 60

After 10 yrs 0.6 20 0.225 70

After 20 yrs 0.525 30 0.15 80

After 50 yrs 0.525 30 0 100

After 100 yrs 0.525 30 0 100

Summary Within this base level sub-catchment the assets most at risk are environmental assets

including Victorian Rare of Threatened Species (VROT), roads and agricultural land therefore

the modelling targeted the protection of these assets. Most of the salinity present in this sub-

catchment is the north-west corner where the Konongwootong Reservoir is situated.

It is evident from the FLOWTUBE modelling that recharge reduction across the entire

modelled area has a significant impact on water table elevation throughout, however,

implication implicating recharge reduction in the top quarter of the catchment has very little

down gradient benefits. However, within this region there may exist a Victorian rare or

threatened species (VROT) that require no significant changes to the hydrology of the

landscape to keep their habitat. In such a scenario, reducing the water table elevations may

cause a change in their habitat so it is necessary to keep this in mind when undertaking any

management actions.

Figure 44 below shows graphically the relationship between recharge reduction and surface

discharge.

64


65

PROCESS OF SETTING ASPIRATIONAL, RESOURCE AND MANAGEMENT ACTION TARGETS Setting targets Targets have been defined and established within the GHCMA Salinity Plan. Their main aim is to

measure the progression toward achievement of their associated goals as required by State

government. Three levels of targets will be set: Aspirational, Resource Condition and Management

Action.

Aspirational Targets Aspirational targets as defined by the GHCMA Salinity Plan (GHCMAa, 2003), are a

statement about the desired condition of the region in relation to salinity in the longer term (50

years +). Actions within the plan have been developed with this long-term goal in mind and

will progressively move the region towards achieving this goal.

Resource Condition Targets

Resource Condition Targets also often referred to as End of Valley Targets provide specific,

time bound, measurable targets for the medium term (10-20 years). The National Framework

for natural resource management standards and targets identifies a minimum set of eight

matters for which targets must be set in the region. Three of these relate to salinity

management.

• Area of land threatened by shallow or rising watertables;

• Surface water salinity; and

• Extent of critical assets identified and protected from salinity and degrading water

quality.

Establishment of appropriate targets requires prediction of trends, an assessment of risk to

assets and values and agreement on the acceptable level of risk. Insufficient data currently

exists in the region to enable appropriate targets to be set for all matters except surface water

salinity. High priority actions have been identified in the GHCMA Salinity Plan (GHCMAa,

2003) to support development of appropriate targets in consultation with the community by

December 2004.

Aspirational Goal: That surface and groundwater salinity levels do not negatively impact on key regional assets.

66

Interim surface water salinity targets have been set for four catchment points by the GHCMA

Salinity Plan (GHCMAa, 2003). Targets indicate the maximum stream salinity level desired

for these rivers by 2012.

• Hopkins River at Wickliffe 15000 EC 90 % of the time;

• Hopkins River at Hopkins Falls 7500 EC 90 % of the time;

• Glenelg River at Sandford 3300 EC 90 % of the time; and

• Wannon River at Henty 5840 EC 90 % of the time.

These targets will be reviewed and may change dependant on our understanding of impact of

the on-ground works in altering the resource condition. Information that will enable us make

this assessment will come from modelling land use change and observing the groundwater

response (eg the impact of trees or pastures on salinity). Extrapolation to broader catchment

scale, and the extended timeframe over which actions may become effective or have an

impact, add additional complexity.

Management Action Targets The GHCMA Salinity Plan (GHCMA a, 2003) identifies practical, achievable actions in five

programs, which contribute to achieving the resource condition targets and our aspirational

goal. These programs are:

• Land Management Program;

• Capacity Building Program;

• Research and Investigation Program;

• Monitoring Program; and

• Coordination Program.

The on-ground works actions of the Land Management Program have specific targets

associated with them. Tree and lucerne programs have been accelerated with 50% of the

required works to be implemented in the first 10 years with the remaining 50% to be

implemented second 20-year period. Targets are reported in Table 34.

67

Table 34. Land Management Program on-ground works targets (after GHCMAa, 2003).

Land Management Program Actions Groundwater Flow Systems Program Action

No. Priority Areas 30yr Target Annual Target

1.1.3, 1.1.5 A1 9273 3091.1.3, 1.1.5 A2 8785 2931.1.4, 1.1.6 B1 418 141.1.4, 1.1.6 B2 737 24

1.1.1 A1 1890 631.1.1 A2 1245 411.1.2 B1 107 41.1.2 B2 125 4

Perrenial Pasture (ha)

Woorndoo, Fractured Paleozoic, Deeply

Weathered Paleozoic1.3.1 A1 28479 949

Discharge Revegetation (ha)

Fencing Discharge (km)

All Systems

All Systems

Accelerated Actions Groundwater Flow Systems Program Action No. Priority Areas 30yr Target Annual Target

(yrs 1-10)Annual Target

(yrs 11-30)

1.3.1 A1 813 41 20

1.1.2 B1 43 2 1.2

Lucern (ha) Woorndoo, Deeply Weathered Paleozoic 1.3.2 A1 1452 73 36

20.3

1

315 158

10

0.6

A1

A1

B1

6310

406

22

Fencing Tree Belts (km)

Western Dundas Tablelands, Eastern

Dundas Tablelands, Deeply weathered granite, fractured

granite, Woorndoo

1.2.3

1.2.1

1.2.2

Tree Blocks (ha)

Fractured Paleozoics, Deeply weathered granite, fractured granite, Merino

Tablelands, Pliocene Sands

Tree Belts (km)

Western Dundas Tablelands, Eastern

Dundas Tablelands, Deeply weathered granite, fractured

granite, Woorndoo

Statistical Analysis Attempts were made to analyse stream salinity trends at each stream gauging stations within

the GHCMA however, long term stream salinity monitoring has been infrequent and irregular

across most of the CMA, making the establishment of salinity trends difficult in many cases.

For each station, all available EC data were used to determine the trends. Obvious outliers,

caused be erroneous flow or EC values, were identified by high residual values and were

removed from the data set and the trends recalculated. There were never more than six

points per station where this was necessary

A graphical summary of the stream EC trend is across the GHCMA is shown in Figure 45.

68

238206

237207

238220 236215

236216

238208

238231

238218

238224

238223

238202 238228238204

236202

236209

237205

237200

Significantly rising EC trend

Non-significant EC trend

Significantly falling EC trend

238206

237207

238220 236215

236216

238208

238231

238218

238224

238223

238202 238228238204

236202

236209

237205

237200







Figure 45. Summary of the statistical analysis of smoothed EC at the major gauging stations within the GHCMA.

Figure 46 to Figure 48 represent smoothed EC, future trends and exceedence curves for the

Hopkins River at Hopkins Falls where the RCT is defined at 7500 EC 90 % of the time.


Hopkins River at Wickliffe where the RCT is defined at 15000 EC 90 % of the time.


Glenelg River at Sandford where the RCT is defined at 3300 EC 90 % of the time


Wannon River at Henty where the RCT is defined at 5840 EC 90 % of the time

69

Hopkins River at Hopkins Falls

Smoothed EC - Hopkins River @ Hopkins Falls

(236209)

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

1973 1979 1984 1990 1995 2001Date

Smoo

thed

EC

(µS/

cm)

Smoothed EC trendMean monthly EC

Figure 46. Statistical analysis of smoothed saltload at Hopkins River at Hopkins Falls.

Smoothed EC - Hopkins River @ Hopkins Falls (236209)

3000

3500

4000

4500

5000

5500

6000

6500

7000

7500

8000

1973 1980 1987 1994 2001 2008 2014 2021 2028 2035Date

Smoo

thed

EC

(µS/

cm)

Min predicted EC (4257 µS/cm) based on current trends

Max predicted EC (4707 µS/cm)based on current trends

Defined RCT(7500 µS/cm)

Future EC range under no change scenario

Defined RCT as stated in the SAP. EC must be

contained below this value 90% of the time.

Figure 47. Smoothed EC trend with future maximum and minimum values by 2012.

70

Exceedence Curve - Hopkins River @ Hopkins Falls (236209)

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

0 10 20 30 40 50 60 70 80 90 100

% of time

EC (µ

S/cm

)


New RCT(7000 µS/cm)

Figure 48. Exceedence curves for Hopkins River at Hopkins Falls [the defined and suggested RCT are also indicated on the graph].

For Hopkins River at Hopkins Falls the defined resource condition target is set at 7500

µS/cm. From Figure 48 we can see that this value is currently being exceeded only 4% of the

time. However we also know that the EC trend is slowly rising 29.6 ± 23.7 µS/cm/yr.

Therefore whilst 7500 µS/cm is an achievable RCT, a slightly more optimistic and achievable

RCT would be 7000 µS/cm by 2012.

71

Hopkins River at Wickliffe

Figure 49. Statistical analysis of smoothed EC at Hopkins River at Wickliffe.

Smoothed EC - Hopkins River @ Wickliffe (236202)

6000

8000

10000

12000

14000

16000

18000

20000

22000

24000

1973 1980 1987 1994 2001 2008 2014 2021 2028 2035

Date

Smoo

thed

EC

(µS/

cm)

Defined RCT(15000 µS/cm).

EC must be contained below this value 90% of

the time.


Min predicted EC (12114 µS/cm)based on current trends



72

Exceedence Curve - Hopkins River @ Wickliffe (236202)

0

2000

4000

6000

8000

10000

12000

14000

16000

0 10 20 30 40 50 60 70 80 90 100

% of time

EC (µ

S/cm

)


Figure 51. Exceedence curves for Hopkins River at Wickliffe [the defined and suggested RCT are also indicated on the graph].

For Hopkins River at Wickliffe the defined resource condition target is set at 15000 µS/cm.

From Figure 51 we can see that this value is currently being exceeded only 5% of the time.

However we also know that the EC trend is slowly rising 306.0 ± 204.1 µS/cm/yr but this trend

is based on a period of data from 1973 to 1987 and extrapolated forward. Therefore, due to

the lack of data the defined RCT of 7500 µS/cm is considered achievable by 2012.

73

Glenelg River at Sandford

Smoothed EC - Glenelg River @ Sandford (238202)

0

2000

4000

6000

8000

10000

1973 1979 1984 1990 1995 2001

Date

Smoo

thed

EC

(µS/

cm)


Figure 52. Statistical analysis of smoothed EC at Glenelg River at Sandford.

Smoothed EC - Glenelg River @ Sandford (238202)

3000

3100

3200

3300

3400

3500

3600

3700

3800

3900

4000

1973 1980 1987 1994 2001 2008 2014 2021 2028 2035

Date

Smoo

thed

EC

(µS/

cm)








74

Exceedence Curve - Glenelg River @ Sandford(238202)

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

0 10 20 30 40 50 60 70 80 90 100

% of time

EC (µ

S/cm

)



Figure 54. Exceedence curves for Glenelg River at Sandford. The Defined and suggested RCT are also indicated on the graph.

For Glenelg River at Sandford the defined resource condition target is set at 3300 µS/cm.

From Figure 54 we can see that this value is currently being exceeded ~60% of the time.

Figure 53 shows that the EC trend is decreasing at -6.8 ± 13.5 µS/cm/yr however it is not

statistically significant. It is therefore considered that current defined RCT of 3300 µS/cm is

not achievable by 2012 and a new RCT of ~6500 µS/cm is a more achievable RCT.

75

Wannon River at Henty

Smoothed EC - Wannon River @ Henty (238228)

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

1973 1979 1984 1990 1995 2001

Date

Smoo

thed

EC

(µS/

cm)


Figure 55. Statistical analysis of smoothed EC at Wannon River at Henty.

Smoothed EC - Wannon River @ Henty (238228)

3000

3500

4000

4500

5000

5500

6000

1973 1980 1987 1994 2001 2008 2014 2021 2028 2035

Date

Smoo

thed

EC

(µS/

cm)








76

Exceedence Curve - Wannon River @ Henty (238228)

0

1000

2000

3000

4000

5000

6000

7000

8000

9000

0 10 20 30 40 50 60 70 80 90 100

% of time

EC (µ

S/cm

)



Figure 57. Exceedence curves for Wannon River at Henty. The Defined and suggested RCT are also indicated on the graph.

For Glenelg River at Sandford the defined resource condition target is set at 7840 µS/cm.

From Figure 57 we can see that this value is currently being exceeded only ~1% of the time.

Figure 56 shows that the EC trend is decreasing at -7.2 ± 15.5 µS/cm/yr however it is not

statistically significant. It is therefore considered that current defined RCT of 7840 µS/cm is

certainly achievable but not very optimistic or aspirational. Setting the new RCT to ~6000

µS/cm is considered more aspirational and achievable RCT.

77

REFERENCES

Cleveland, W.S. (1994). ‘The Elements of Graphing Data.’, 297 pp. (Hobart Press: New

Jersey.)

Cox, J., Fitzpatrick, R., Mintern, L., Bourne, J., and Whipp, G., (1999). Managing waterlogged and

saline catchments in south-west Victoria. A soil-landscape and vegetation key with on-farm

management options. Woorndoo Land Protection Group Area Case Study. Catchment

Management Series, Report Number 2.

Dahlhaus, P., Heislers, D., and Dyson, P., (2002). Glenelg Hopkins Catchment Management

Authority, Groundwater Flow Systems. Consultancy report no. GHCMA 02/02, Dahlhaus

Environmental Geology Pty Ltd.

Dahlhaus P.G., MacEwan R.J., Nathan E.L. & Morand V.J. 2000. Salinity on the southeastern

Dundas Tableland, Victoria. Australian Journal of Earth Sciences 47/1 pp. 3-11

Dahlhaus P.G. & MacEwan R.J. 1997. Dryland Salinity in South West Victoria - Challenging

The Myth. in (G.McNally, ed.) Collected Case Studies in Engineering Geology,

Environmental Geology, Hydrogeology, Third Series, Geological Society of Australia

Inc. pp. 165-181

Dawes WR, Walker GR, and Stauffacher M (1997) Model Building: Process and Practicality.

In: Proceedings of MODSIM 97, McDonald AD, McAleer M (eds). Modelling and

Simulation Society of Australia: Canberra; 317-322.

Dawes, W.R., Stauffacher, M. and Walker G.R. (2000) Calibration and Modelling of

Groundwater Processes in The Liverpool Plains. Technical Report 5/00, CSIRO Land

and Water, Canberra.

Dawes W., Gilfedder M., Walker G.R. and Evans W.R. (2001) Biophysical modelling of

catchment scale surface and groundwater response to land-use change. In F.

Ghassemi, P. Whetton, R. Little, and M. Littleboy (Eds), Proceedings of MODSIM 2001.

Modelling and Simulation Society of Australia and New Zealand, Canberra, 535–540.

GHCMA a, (2003). Glenelg Hopkins CMA Salinity Plan (Revised Draft).

Dixon P.1989. Dryland Salinity in a subcatchment at Glenthompson, Victoria. Australian

Geographer 20(2), 144 – 152

Consultancy report by Department of Primary Industries.

GHCMA b, (2003). Glenelg Hopkins Regional Catchment Strategy 2003 – 2007.

Commonwealth publication for National Action Plan for salinity and Water

Quality.

Hastie, T.J. and Tibshirani, R.J. (1990). ‘Generalized Additive Models.’ (Chapman and Hall:

New York.).

Heislers, D. and Brewin, D. 2003. Salinity prioritisation in the Glenelg Hopkins region.

Background report for the Glenelg Hopkins Salinity Plan. CLPR Research Report No,

23 June 2003.

78

Hirsch, R.M., Slack, J.R., and Smith, R.A. (1982). Techniques of trend analysis for monthly

water quality data. Water Resources Research 18, 107-21.

Jolly, I.D., Williamson, D.R., Gilfedder, M., Walker, G.R., Morton, R., Robinson, G., Jones, H.,

Zhang, L., Dowling, T.I., Dyce, P., Nathan, R.J., Nandakumar, N., Clarke, R., and

McNeill, V. (2001). Historical stream salinity trends and catchment salt balances in the

Murray-Darling Basin, Australia. Marine Freshwater Research 52, 53-63.

Morton, R. (1997). Semi-parametric models for trends in stream salinity. Report Number CMIS

97/71 (CSIRO Mathematical and Information Sciences: Canberra.)

Nathan E.L. 1999. Dryland Salinity on the Dundas Tableland: a historical appraisal.

Australian Geographer, 30/3 pp.295-310.

Paine, M. D., Bennetts, D. A., Webb, J. A. & Morand, V. J. 2004. Nature and extent of Pliocene

strandlines in southwestern Victoria and their application to late Neogene tectonics.

Australian Journal of Earth Sciences, (in press).

Smitt, C., Dahlhaus, P. and Cox, J. 2003. Setting aspirational, achievable and management action

targets across the Glenelg Hopkins CMA. 1st Milestone Report - Review of the Groundwater

Flow Systems and analysis of available surface and groundwater data. CSIRO Land and

Water Client Report.

Stuart-Smith, P. G. & Black, L. P. 1999. Willaura sheet 7422, Victoria,

1:100,000 map geological report. Australian Geological Survey Organisation Record,

1999/38.

setting aspirational, resource and management action ... · approach) regression. if...

Documents