BookPDF Available

Healthy Waterways Management Strategy Evaluation: Science support for catchment-to-coast water quality management

Authors:
Report to the Healthy Waterways Partnership
October 2008
Francis J. Pantus Eva Abal
Leonie Pearson Andy Steven
Healthy Waterways Management
Strategy Evaluation: Science support for
catchment-to-coast water quality management
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
A Wealth from Oceans Flagship report, ISBN: 978-1-921424-68-7
On-line version: ISBN: 978-1-921424-69-4
CD-ROM version: ISBN: 978-1-921424-70-0
Australia is founding its future on science and innovation. Its national science agency, CSIRO, is a
powerhouse of ideas, technologies and skills.
CSIRO initiated the National Research Flagships to address Australia’s major research challenges
and opportunities. They apply large scale, long term, multidisciplinary science and aim for
widespread adoption of solutions. The Flagship Collaboration Fund supports the best and brightest
researchers to address these complex challenges through partnerships between CSIRO,
universities, research agencies and industry.
The Wealth from Oceans Flagship, together with its research partners, is providing Australia with a
key capacity to discover, protect and realise the benefits of our ocean territories. The work
contained in this report is done in collaboration between CSIRO National Research Flagships and
The Healthy Waterways Partnership, Queensland.
For more information about Wealth from Oceans Flagship or the National Research Flagship
Initiative visit http://www.csiro.au/org/WealthOceansFlagship.html.
For more information about the SEQ Healthy Waterways Partnership, visit
http://www.heatlhywaterways.org.
Citation: Pantus Fancis J., Abal E, Pearson L, Steven A, (2008). Healthy Waterways Management
Strategy Evaluation: science support for catchment-to-coast water quality management. CSIRO:
Wealth from Oceans and Water for a Healthy Country National Research Flagships. CSIRO
Publishing, Cleveland, Australia. 81 pp.
Copyright and Disclaimer
© 2008 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.
Important Disclaimer:
CSIRO 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 (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 Photograph:
Description: SE-Queensland is one of Australia’s fastest growing regions. Catchment-to-Coast
Management Strategy Evaluation (C2C MSE) is new science initiative that aims to support all major
elements of (water-related) adaptive natural resource management. SE-Queensland was chosen as
the preferred region for a pilot study due to its already well developed coastal management
processes, available data and wide-ranging collaborations between stakeholders. Report cover:
Louise Bell and Francis Pantus, map graphics courtesy of SEQ-HWP.
i
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Healthy Waterways
Management Strategy Evaluation
Science support for
catchment-to-coast water quality management
ii
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
iii
PREFACE
This document outlines the results of a one-year project to develop a ‘proof of concept’ of a Healthy
Waterways Management Strategy Evaluation (HW MSE) system which makes operational the
adaptive management framework in management of waterways in South East Queensland. The
development of the HW MSE is based on the application of the Catchment-to-Coast MSE (C2C-
MSE) concepts. This report is also submitted to the SEQ Healthy Waterways Partnership as the
final deliverable of Phase 1.
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
ACKNOWLEDGEMENTS
The project is a collaborative initiative of CSIRO and the SEQ Healthy Waterways Partnership.
We would like to thank and acknowledge all individuals and organisations that supported us in
making this project possible, especially:
Healthy Waterways Scientific Expert Panel (chaired by Prof. Paul Greenfield)
Healthy Waterways Technical Advisory Group
Healthy Waterways Partnership Secretariat (led by Di Tarte)
BMT WBM (Tony McAlister and Tony Weber)
E-water CRC (Joel Rahman and Geoffrey Davis)
CSIRO Wealth from Ocean Flagship (Drs Kate Wilson and Bill de la Mare)
CSIRO reviewers Drs Cathy Dichmont and Sean Pascoe
A special thanks to Toni Canard for proof-reading and editing the final document.
iv
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
EXECUTIVE SUMMARY
The management of our water resources is complex with many and often conflicting uses. To help
address regional water-related issues, the SEQ Healthy Waterways Partnership (HW-Partnership)
was established in 2001 (http://www.healthywaterways.org/). The HW-Partnership is a special
collaboration between government, industry, researchers and the community. These Partners work
together to improve catchment management and waterway health in the eastward-draining rivers in
SE-Queensland, Australia. One of the products of this collaboration is the Healthy Waterways
Strategy 2007 – 2012. The Strategy comprises of an integrated set of Action Plans, designed to
maintain and improve the health of the waterways and catchments of South East Queensland. The
Management Strategy Evaluation (MSE) program is part of the Enabling Action Plan in the Strategy.
The MSE Action Plan aims to help achieve the SEQ Healthy Waterways Partnership vision by
underpinning it by an integrated knowledge and information, modelling and monitoring framework in
the context of an adaptive management approach.
Management Strategy Evaluation is an approach to support natural resource management with a set
of concepts, standards and outputs that allows policies and management scenarios to be evaluated
for their impacts on social, environmental and economic values.
One of the key capabilities of the application of MSE is to deliver systematic ‘what-if’ scenario
evaluation functionality, based on the best available scientific information. To examine the value of
MSE for SE-Queensland, a one year demonstration-of-concepts project was approved.
The brief of the pilot project was to examine the success and challenges of MSE for water quality
management in SE-Queensland and, if appropriate, scope a follow up project for the SE-QLD region.
This report describes the work and results of this project. The emphasis of this report is not on
performing a full-scale MSE for an area within the SE-Queensland region, but to look at the potential
of the MSE method to support complex natural resource management.
This document reports our findings of designing and implementing a pilot MSE and a glance into the
future. This report is not a comparative study of alternative management support systems, nor does
it give us reliable results on management options for SE-Queensland water quality management.
Setting some boundaries
When starting the Healthy Waterways MSE (HW-MSE) project, some choices needed to be made
regarding the geographic area of interest, the processes to be included in representing the expected
ecosystem responses to management actions and the management levers to be implemented.
Catchments are relatively well-defined geographic systems with respect to water so they are a
natural boundary to delineate the region. The choice for the Logan-Albert catchment as the focus
area for the pilot MSE was made based on its expected rapid population expansion, diversity of
issues and availability of a calibrated catchment model (hydro-dynamics and transport). The Logan-
Albert was also explicitly mentioned in the Strategy 2007 – 2012 document as one of the
‘deteriorating catchments’.
The main driver for change in the Logan/Albert catchment is expected to be the shift in land-use
caused by the planned demographic changes over the next 15 years. Land-use is one of the levers
of management in the catchment and land-use is also an important variable in determining the runoff
of water and constituents (sediments, nutrients). This meant that the response model chosen to
represent the catchment’s responses to management was structured around land-uses and runoffs.
To show the implications of management to other than biophysical processes, a model of the
aquaculture industry around the mouth of the Logan/Albert was also developed. This model extends
the performance biophysical measures with economic and social (employment) measures.
Apart from managing land-use, the second class of management levers that were implemented is a
set of remedial actions. Examples of such important remedial actions as defined in the Strategy
2007 – 2012 are restoration of riparian vegetation and water-sensitive urban design. To simulate the
effects of such remedial management actions, a general mechanism (borrowed from the EMSS
model suite) was implemented to allow the user to moderate the potential flows and constituent
loads from a land-use into the waterways, thus simulating the effects of such management actions.
v
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
As the effect of a given management action may depend on the amount of flow or constituent load,
the MSE allows the specification of relationships for water, sediment and nutrient runoff between a
land-use and receiving waters.
To demonstrate the use of the MSE approach, we set ourselves as a goal to implement key MSE
standards:
1. Deliver capability to evaluate systematically multiple-use management scenarios by
presenting a set performance indicators that allows consideration of trade-offs between
those scenarios.
2. Incorporate all major elements of ‘real world’ adaptive (multiple use) management, including
their restrictions and dynamic feedback. We decided on six elements: management
decision, management action, system response, observation, assessment and learning.
3. Explicitly deal with the uncertainty in models and outputs.
4. Reports the trade-offs between management scenarios, not just single solutions
5. Integrate many management activities and tasks
MSE is a complex concept, and to make its functionality for resource management more tangible, we
developed a (pilot) software application around the Logan/Albert catchment, implementing five MSE
standards. This approach has a couple of advantages: (i) it gives us a product to actually show the
working and results of MSE to our stakeholders, (ii) it teaches us how to structure ‘step-by-step’
support for designing and evaluating management scenarios, and (iii) it informs us on the challenges
of the software implementation of such a complex application.
A feature of the pilot MSE software is that it allows us to remain flexible with respect to the many
models (and their parameterisation) that make up a MSE. Analogous to computer hardware, we
developed a novel software architecture that supports model ‘plug-and-play’ functionality. To better
demonstrate the workings of a MSE application, an intuitive user interface that supports the workflow
arising when performing a MSE (definition, evaluation and analysis of management scenarios), has
been developed and implemented.
The current software supports all elements of an MSE: it allows the user to define and evaluate
scenarios using two levels of tools (general parameter editing and model-specific editing), it allows
fairly extensive monitoring of the evaluation process itself and it allows some level of analysis of
results and moving between high-level and low-level results. It allows a broad range of management
scenarios to be handled with relative ease. The software application demonstrates how the MSE
principles can be used to evaluate coastal water-resource management scenarios in an informative
and consistent way.
However, at the completion of the pilot project, the learning and decision making elements do not
contain the relevant functionality yet, and the dynamic adaptive feedback needs to be implemented
manually. The incorporation of suitable management decision and learning strategies is one of the
high-priority tasks identified and the appropriate models for the learning and decision making are
currently under development.
MSE as an integration platform
The pilot MSE application demonstrates the ability of MSE to work as a point of integration for a
range of processes. The MSE application integrates management elements such as observation,
assessments, and management actions. The application also allows the adaptation of each one of
those elements to better describe the system under consideration. This allows us to bring together
elements of existing programs such as the Ecosystem Health Monitoring Program and the annual
report card assessment. Together with the ability of the MSE application to apply management
actions such as land-use and remediation scenarios (water-sensitive urban design, riparian
revegetation), the pilot application allows us to evaluate a range of management actions proposed in
the Strategy 2007 – 2012.
vi
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
The MSE application also brings together the various models that describe our understanding of the
managed system and its responses to the management actions (catchment physics model,
aquaculture socio-economics model). The pilot application demonstrates the principles and pitfalls of
integrating component models (e.g. catchment physics and aquaculture economics) into complex
response models in a flexible fashion. Especially this part of the application needs more work to
provide better support for users by combining component models and allow checking of the validity
of resulting composite response models.
Challenges
Part of the learning in HW-pilot MSE is that we now have a better understanding of the challenges
that we’re facing in developing the concepts and implementation of catchment-to-coast MSE. Some
of the more fundamental challenges are around the integration of our knowledge and abilities to
support resource management.
Conceptual development
Understanding and depicting the place of water in the complex network of processes and
dependencies in SE-Queensland will be a challenge. Such understanding may need to include
many cross-links between the physical, biological, economic, social and even political realms. As
SE-Queensland is getting to grips with the large demographic changes over the next ten years,
possibly exacerbated by climate change effects, such understanding has to keep pace with the
changes in the use and management of our water resources.
Highly connected and non-linear systems are complex systems, in the scientific meaning of the term.
Diminished predictability of the system’s temporal behaviour and effectiveness of management may,
among others, result from such complexity. The challenge is to understand the implications of such
complexity regarding the development of our system understanding and management. Such insights
may even change our thinking on how to manage our resources effectively.
Science integration
Our science is often arranged along discipline and institutional lines. For science to be even a
stronger ally in whole-of-water cycle management underpinned by the catchment-to-coast
philosophy, it needs to be delivered in an integrated way, as promoted by the MSE approach.
Breaking through the walls that often separate the various strands of science, and forge them into an
integral body of knowledge is critical in the success of the HW MSE development. Using an
integration framework such as MSE to focus our activities is a crucial step forward.
Science priorities need to be developed in close collaboration between resource management and
science practitioners. We expect for instance that the decision-making and learning elements of
MSE fall into this class. To sustain such collaborations over many years (or even decades) will be a
challenge.
Implementation and acceptance
At its core, the MSE approach promotes integration of all major elements of an adaptive
management system. The mechanics, implementation and daily management of such integration
faces challenges ranging from available expertise to long-term funding. Deliverables for a range of
areas (e.g. science, monitoring, management) need to be developed from the start with such
integration in mind. One of the challenges is to develop a range of support mechanisms and
standards to move towards the realisation and acceptance of the need for integration with
stakeholders and science providers.
As SE-Queensland heads towards integrated water resource management, we need to start the
discussion on how to best meet such challenges. Various avenues/mechanisms to facilitate this
discussion and subsequent collaborative efforts need to be scoped. The development and
operationalisation of the HW MSE philosophy is very dependent on such mechanisms.
vii
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Where from here for Healthy Waterways MSE?
After having concluding the pilot project successfully, the project brief specifies the delivery of a
vision for follow-on projects for the mid-term (five-year) future. Main areas of development are: (i)
functional extension of the MSE application, (ii) development of science methods, (iii) development
and implementation of technical capability into the MSE application, and (iv) extension to make MSE
concepts and functionality more accessible to a wider group of stakeholders.
Functional extension of the MSE
Extending the functionality of the HW-MSE would include adding more catchments to the current
MSE and validating their working. The two catchments earmarked for inclusion are the Brisbane and
Maroochy sub-catchments. Extending the range of models that deal with important issues such as
costing of management actions and socio-economic responses (or management levers) are also
high on the agenda and are of direct interest to our stakeholders.
Science development for the HW-MSE
Science development is necessary to enable us to extend the existing MSE functionality. More
conceptual development is needed to integrate biophysics, economics and social knowledge,
management levers and performance measures within the MSE framework. Research on how
ecosystem services can help us to put more emphasis on the economic aspects of resource
management is expected to be a part of the future science program. Preliminary thoughts on how to
include indicators for institutional arrangements in the next iteration of the HW-MSE approach have
been discussed and already a program of monitoring key economics for SE-Queensland is
underway. Research regarding scientific methods that allow us to test options for learning and
decision making in a complex, data-poor but ADAPTIVE environment have already commenced.
The development of a better or fully integrated catchment-to-coast biophysical model will allow us to
better understand the impacts of management actions on the three main sections: fresh water,
estuarine and the receiving marine environment. Synergies between the current needs for improved
model capability of our stakeholders in the estuarine and marine section and the MSE need for
better integrated biophysical models may create an opportunity for such development. Science
development is also needed to support better understanding (statistical, visual) the high-dimensional
results of complex management scenario evaluations.
Technical development of HW-MSE application
Technical development of the MSE application is expected to build on the existing (plug-and-play)
software architecture. Of high priority are software modules that better support the configuration of
additional areas and catchments. A better user interface that supports constructing complex
response models from a range of models that implement our system understanding is also high on
the agenda. The structure of MSE stochastic evaluation lends itself to ‘parallel processing and
distributed computing (many computers work together to finish a task) is expected to become a
necessity to run the HW-MSE. As the HW-MSE grows more complex by adding more catchments,
more response models, more performance measures, more management actions, more
performance indicators etc., keeping track of the results of the management scenarios needs
special attention. Capabilities to store large amounts of data will need powerful database storage
capacity. To retrieve information in a way that allows moving between different levels of resolution
needs data warehousing capabilities such as hypercube technology. To manipulate results involving
many variables from high-dimensional domains (e.g. scenarios, models, variables, and space-time)
and make them accessible in an informative way to end-users, needs a high level of functionality on
the part of the software implementation.
To make the HW-MSE available to a wider range of stakeholders (see next section), web-based
technologies such as (secure and restricted) internet access to MSE scenario definition and results
are expected to be employed. Some initial enquiries into the feasibility of such interfaces are already
underway.
Extension of HW-MSE
Without a strong base in the application of the HW-MSE to real-world resource management issues
in SE-Queensland, the long term and sustained development of something as complex as
Catchment-to-Coast MSE is expected to wither. This means that the collaboration between resource
viii
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
ix
managers and science providers needs to be continued and strengthened. We see this
collaboration as one of the most important success factors for the further development of the HW-
MSE and there are already proposals to strengthen the links with our stakeholders through the
development of training materials and documentation. We also expect a synergetic exchange
between the experiences of the real-world management and the learning from the MSE, thus
creating another adaptive loop.
Extension with the scientific community is another link that needs our attention. The standard
communication tools of publication in peer-reviewed journals and presentation at national and
international scientific conferences are part of keeping the science community informed. We expect
that the number of collaborative science projects will further increase based on the needs for
innovative science by the SE-Queensland stakeholders.
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
CONTENTS
1. Introduction........................................................................................................................1
1.1. Rationale ...................................................................................................................1
1.2. SEQ context ..............................................................................................................3
1.3. Project brief and outputs ...........................................................................................5
2. Management Strategy Evaluation....................................................................................7
2.1. Management Scenario and Strategy Evaluation.......................................................7
2.2. Scoping an MSE application .....................................................................................9
2.3. The MSE elements..................................................................................................11
2.4. Other functions of MSE ...........................................................................................13
2.5. Summary.................................................................................................................13
3. Catchment-to-coast MSE in SE-Queensland................................................................15
3.1. SE-Queensland water management issues............................................................15
3.2. Applying an MSE approach to SE-Queensland water issues.................................16
3.3. MSE as a platform for integration............................................................................17
3.4. The Logan-Albert demonstration project.................................................................20
3.5. The Logan-Albert catchment...................................................................................22
3.5.1. Logan/Albert overall statistics ............................................................................. 24
3.6. Summary.................................................................................................................25
4. Logan/Albert pilot mse models......................................................................................27
4.1. The catchment model..............................................................................................27
4.2. The industry models................................................................................................35
4.2.1. Aquaculture......................................................................................................... 35
4.3. The observation model............................................................................................37
4.4. The assessment model...........................................................................................37
4.5. Institutional arrangements.......................................................................................38
4.6. The management actions model.............................................................................41
4.7. The Management Decision and Learning models...................................................42
4.8. Summary.................................................................................................................43
5. The Logan-Albert pilot mse SOFTWARE......................................................................45
5.1. HW-MSE application preliminaries..........................................................................45
5.2. HW-MSE application workflow................................................................................45
5.2.1. Specification ....................................................................................................... 45
5.2.2. Scenario Evaluation............................................................................................ 48
5.2.3. Monitoring the evaluation.................................................................................... 48
5.2.4. Results: the management decision support table ............................................... 49
5.2.5. Post-processing.................................................................................................. 50
5.3. A session with the Logan/Albert HW-MSE software...............................................51
5.3.1. Example 1, coastal urban development with and without remediation................ 51
5.3.2. Example 2: the influence of the monitoring interval ............................................ 54
5.4. Summary.................................................................................................................59
6. Conclusion and future ....................................................................................................61
6.1. Discussion and conclusion......................................................................................61
6.2. Future of C2C MSE.................................................................................................62
6.3. Summary.................................................................................................................63
Appendix A – software development.......................................................................................65
Appendix B –The E2/HW-MSE Catchment model...................................................................72
Appendix C: Glossary ...............................................................................................................77
x
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
LIST OF FIGURES
Figure 1-1 South-East Queensland (red box) covers 2,300 km2, 1.3% of the mainland area of
Queensland and is home to more than 50% of the population of Queensland. ...........................3
Figure 1-2 South-East-Queensland extends from the NSW border in the south to Noosa in the north
and from the Stradbroke Island in the east to the reaches of the Brisbane river in the Great Dividing
Range in the west. The area spans about more than 22,600 km2 and includes the catchments of 14
major rivers.................................................................................................................................... 4
Figure 1-3 The adaptive management cycle is a powerful ally in managing systems that contain
many uncertainties. Graphics courtesy Keith Sainsbury..............................................................5
Figure 2-1 The six elements of an adaptive management system............................................... 9
Figure 2-2 MSE interacts with the planning, evaluation and learning and adjustment elements of the
adaptive management cycle........................................................................................................10
Figure 2-3 An MSE simulates the essence of an adaptive resource management system........11
Figure 3-1 The collective mind-map of experts in SE-Queensland regarding the region’s water
issues. .........................................................................................................................................15
Figure 3-2 SE-Queensland has a range of water-related management issues (graphics courtesy of
the SEQ-HWP)............................................................................................................................16
Figure 3-3 A range of activities and information is used in the management of SE-Queensland
waterways....................................................................................................................................18
Figure 3-4 MSE as an integration platform: many activities (and information) find a place in the MSE
framework. The MSE framework effectively connects and thereby integrates these activities..19
Figure 3-5 Sketching the MSE as a conceptual diagram is a useful first step in designing a MSE
application. We used IAN (http://ian.umces.edu/symbols) symbols to construct this graphic.... 21
Figure 3-6 The Logan-Albert catchment is one of Queensland's most southern catchment. Its main
rivers are the Logan and the Albert (southern branch on the image)......................................... 23
Figure 4-1 A conceptual diagram representing the daily rain runoff model ‘SIMHYD’ (Podger, 2004).
.....................................................................................................................................................28
Figure 4-2 The Logan-Albert E2 catchment model uses 37 sub-catchments to calculate water and
constituents runoff (coloured polygons). The 37 sub-catchments are represented by a directed node
network (the arrows and black dots)...........................................................................................29
Figure 4-3 Potential Evapo-Transpiration or PET values for the Logan-Albert region. The bars
indicate the standard deviation of inter-annual variation of PET values between 1983 and 2003.30
Figure 4-4 The (SILO) daily rainfall data is accumulated into annual averages for the Logan-Albert
catchment. The spatial variability over the catchment is reflected in the standard deviation bars in the
graph. ..........................................................................................................................................31
Figure 4-5 The pilot HW-MSE catchment model is based on the 1999 land-use map, produced by
the State-wide Land-cover and Trees Study (SLATS) program .................................................32
Figure 4-6 The Logan-Albert catchment model’s simulated flow, based on the SILO................ 32
Figure 4-7 The Logan-Albert catchment model results for the three main transport constituents: total
nitrogen, total phosphate and suspended solids.........................................................................34
Figure 4-8 The development of aquaculture in the Logan-Albert catchment. From: Report to Farmers
- Aquaculture production survey, Queensland 1990-91 to 2006-07 (QDPI, 2008)..................... 36
Figure 4-9 A screen shot of the user interface that supports the entry of n management actions.41
Figure 4-10 The remediation activities entry screen.................................................................. 42
Figure 5-1 The specification screen allows creating different models and construct scenarios. 46
xi
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 5-2 Models may have thousands of parameters. Special-purpose screens are available to
support easy access to those parameters and make editing less error-prone............................47
Figure 5-3 A MSE is a collection of scenarios. Each scenario consists of some administrative
information and a range of models, representing the major management elements..................47
Figure 5-4 After composing the management scenarios, the next activity is to evaluate them. 48
Figure 5-5 The Monitor tab allows inspecting the results of various models...............................49
Figure 5-6 A key output of a MSE is the management decision support table. It shows the tradeoffs
between the various management scenarios, expressed in terms of performance measures...49
Figure 5-7 The Analyse tab provides simple but flexible tools to perform more detailed analyses. T
allows analyses on stochastic averages and on the stochastic replicates (shown)....................50
Figure 5-8 The left panel shows the screen to define the urban land-use scenario. The right panel
shows the screen for defining the remedial management actions scenario................................52
Figure 5-9 The left panel show the screen used to combine the models into the management
scenarios. The right panel show the screen that allows us to run the four scenarios................52
Figure 5-10 The time series for total nitrogen concentration (TNConc) for two scenarios, remedial
and status quo, are shown on the graph. ....................................................................................53
Figure 5-11 Dividing two stochastic time series (remedial and status quo TN concentration) allows
easier comparison between the effects of two management scenarios. The green line indicates that
there is no difference between the scenario results until beginning of 1995, and the quotient is
around unity. The red line indicates that the remedial management scenario results in around 4%
reduction in total nitrogen concentration after 1995. ...................................................................53
Figure 5-12 Specifying models is often the first step in composing a scenario...........................55
Figure 5-13 Composing a new scenario is simply a task of choosing from the available models and
saving the result...........................................................................................................................56
Figure 5-14 During the evaluation of the scenarios, there is ample visual feedback given to the user
about the progress and performance of the running models.......................................................57
Figure 5-15 After evaluating the scenarios, we can inspect the results by dragging and dropping time
series from the table (top) onto the graph (bottom).....................................................................57
Figure 5-16 The graph editor delivers a range of function to combine or summarise data series. This
is accomplished by adding a new series that is the result of applying a function, averaging in this
case..............................................................................................................................................58
Figure 5-17 The graph editor allows some basic assessment and comparison of time series. See
text for more details. The graph on the right plots the average of the total nitrogen values against the
observation interval......................................................................................................................58
xii
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
xiii
LIST OF TABLES
Table 3-1 The projected mainland areas of the SE-Queensland catchments............................ 20
Table 3-2 Population growth predictions (x10,000) (original data: Qld Office of Urban Planning)24
Table 3-3 Areas of land-use in the Logan/Albert catchment model............................................24
Table 4-1 The measured long term runoffs, cited from the Australian Natural Resources Atlas.33
Table 4-2 The basis of the aquaculture industry model is a model simulating the economics of a
aquaculture farm in the Logan-Albert region............................................................................... 36
Table 4-3 The reference values used in the SEQRMS assessment model at the Logan/Albert river
mouth...........................................................................................................................................37
Table 4-4 Suggested measures of HW Partnerships vision and objectives for further consideration in
MSE development....................................................................................................................... 40
Table 5-1 Descriptions of the four management scenarios. .......................................................51
Table 5-2 The management decision support table shows the results of the four management
scenarios. The values are the averages +/- the standard deviations, calculated over time after
01/01/1996 and over stochastic replications...............................................................................54
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
1. INTRODUCTION
1.1. Rationale
This report describes the results of a pilot study to demonstrate the application of Management
Strategy Evaluation (MSE) concepts in a Catchment to Coast (C2C) environment. The application of
these concepts to SE-Queensland is called the Healthy Waterways MSE (HW-MSE).
Aquatic ecosystems were deteriorating through the rapidly growing use of natural resources in SE-
Queensland (Australia) in the late 1980’s and early 1990’s. A range of organisations came together to
turn around the decline. These organisations are representative for 18 local councils, traditional
owners, community groups, Queensland State and industry. The organisation they formed is now the
SE-Queensland Healthy Waterways Partnership (SEQHWP) and information on its history,
composition, focus and tasks can be found on line at
http://www.healthywaterways.org/who_is_hww_home.html.
There are a range of activities currently in place in SEQ that would benefit from explicit integration.
One of the action plans in the Strategy is called the Management Strategy Evaluation Action Plan.
This Action Plan has at its core an attempt to develop and test the applicability of the MSE approach to
the complex environment of SE-Queensland water management. It aims to integrate freshwater,
estuarine and marine ecosystems and the multiple management regimes operating in the region, as
well as the human and natural impacts on these systems.
Also, since the Partnership started in the early 1990’s, its many activities have delivered a better
understanding and management of the many parts of the catchments and their waterways. Those
activities also included examining available management levers and assessment methods. An
extensive monitoring program for the fresh water, estuarine and marine domains has been underway
since the early 2000’s. Integrating the results of all these activities (e.g. system understanding,
planned management actions, monitoring and assessment results) poses a challenge at various
levels, from conceptual to operational.
Management Strategy Evaluation (MSE) is an approach to support natural resource management with
a range of concepts, standards and products. Differentiating characteristics of MSE are:
1. Delivery of capability to systematically evaluate multiple-use management scenarios by
presenting a set performance indicators that allows consideration of trade-offs between those
scenarios.
2. Incorporation all major elements of ‘real world’ adaptive (multiple use) management, including
their restrictions and dynamic feedback (management decision, management action, system
response, observation, assessment and learning).
3. Explicitly deal with the uncertainty in models and outputs.
4. Reports the trade-offs between management scenarios, not just single solutions
5. Integrates many management activities and tasks
Especially the integrative nature of the MSE approach and its ability to support adaptive management
is attractive for better science-support in managing SE-Queensland’s waterways. To test and
demonstrate the concepts and workings of a Catchment to Coast MSE (C2C-MSE), the SEQHWP
commissioned a one-year pilot project. The description of this project and the report on its findings are
the subject of this document.
1
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
The report is organised as follows:
Chapter 1 introduces the SE-Queensland region and discusses some of the challenges in managing
the waterways as well as some strategies to cope with those challenges.
Chapter 2 shows the place of MSE in the wider framework of resource management and discusses the
concepts and details of the MSE approach. It emphasises the importance of the Strategy for the
scoping of an MSE and expands the usefulness of MSE beyond its scenario evaluation capabilities.
Chapter 3 discusses the background of SE-Queensland water quality challenges and management
and links the SE-Queensland water issues to the MSE approach. It also introduces the pilot MSE that
has been developed for the Logan-Albert catchment.
Chapter 4 presents the main models implemented into the pilot Healthy Waterways MSE (HW-MSE)
application for the Logan-Albert catchment.
Chapter 5 shows the main workings of the pilot HW-MSE application and shows some results to
demonstrate its application and potential use.
Chapter 6 contains a conclusion with respect to the application of MSE principles to the coastal zone
management and an outline of future work that would be needed in the follow-on phase of this pilot
project. Some of the technical details have been concentrated in the Appendices to make this report
accessible to non-technical readers.
2
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
1.2. SEQ context
South East Queensland is a region on the east coast of Australia (the red box in Figure 1-1), centred
around Brisbane. As Figure 1-2 shows, SE-Queensland extends from Noosa in the north to the New
South Wales border in the south and is the region in Australia with the largest growth: a net increase of
over 64,200 people in the period 2006-2007 (ABS report 3218.0 - Regional Population Growth,
Australia, 2006-07).
Over the next 20 years the population is expected to grow from a total of 2.7 million in mid-2006 to four
million people or more in 2026. This rapid growth will result in increasing demands for reliable supplies
of potable water, increased recreational pressure on natural areas such as Moreton Bay and inland
waterways, and greater demands for roads, housing, shopping centres and industrial estates. Without
careful management, these demands are likely to lead to further degradation of our waterways.
Figure 1-1 South-East Queensland (red box) covers 2,300 km2, 1.3% of the mainland area of
Queensland and is home to more than 50% of the population of Queensland.
The coastal zone, for the purposes of this report, is the strip of land from the western boundaries of the
catchments, including the near-shore coastal waters such as Moreton Bay and the Broadwater. The
coastal zone is a very complex ecosystem which we expect to sustain a multitude of land-uses,
assimilate large amounts of our waste products and still deliver a range of (ecosystem) services (from
drinking water to clean air) without which our lives would be much harder, if not impossible.
The major water quality issues affecting SE-Queensland (Figure 1-2) include excessive levels of
sediment and nutrients entering our waterways from both urban and non-urban catchments, resulting
in reduced natural flows and considerable habitat loss. The sediments and nutrients come from both
point sources and diffuse sources (see glossary at the end of the document). The solutions to these
excessive levels of pollution lie in reducing our current inputs and minimising the impacts of future
population growth.
3
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 1-2 South-East-
Queensland extends from
the NSW border in the
south to Noosa in the
north and from the
Stradbroke Island in the
east to the reaches of
the Brisbane river in the
Great Dividing
Range in the west. The
area spans about more
than 22,600 km2 and
includes the catchments
of 14 major rivers.
Due to these rapidly
increasing demands on
our environment,
there is a strong need
for an enhanced
capacity to make
decisions and undertake
appropriate management
actions for stakeholders
and resource managers.
However, the management
of our natural resources in a
complex system such
as our coastal zone is far
from being completely
understood and mastered,
and we are developing
and testing various approaches in different parts of the world (Carpenter et al. 1999, Carpenter and
Gunderson 2001, Briggs 2003, Berkes, 2006).
One of such approaches is the adaptive management approach (see Figure 1-3), which is based on
the recognition that we often need to act on the basis of an imperfect understanding of the systems
within which management occurs. This approach is based on the thesis that on-going knowledge
development (e.g. through monitoring, evaluation and active learning Walter and Holling 1990) is
needed to improve our understanding and capability to manage such a system.
Aquatic ecosystems were deteriorating through the rapidly growing use of natural resources in SE-
Queensland in the late 1980’s and early 1990’s. This triggered 18 councils, traditional owners,
community groups, the Queensland State, and industry to form an organisation now known as the SE-
Queensland Healthy Waterways Partnership (SEQHWP). One of the key responsibilities of the
Partnership is to deliver a strategy to better manage our aquatic resources, referred to in this report as
the Strategy. The Management Strategy Evaluation Action Plan is one component of the SEQ Healthy
Waterways Strategy and is one of the three enabling action plans contained within the strategy. The
purpose of the MSE Action Plan is to ensure that:
The achievement of the Healthy Waterways vision will be underpinned by an integrated knowledge
and information, modelling and monitoring framework in the context of an adaptive management
approach
This plan, together with all of the other action plans comprising the SEQ Healthy Waterways Strategy,
contributes towards the achievement of the Healthy Waterways vision, which is:
By 2020, our waterways and catchments will be healthy ecosystems supporting the livelihoods and
lifestyles of people in South East Queensland, and will be managed through collaboration between
community, government and industry.
4
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
The aim of management of the coastal zone is to optimise the long-term costs and benefits of our use
of the natural resources to our disposal. The growing awareness within the population that our
ecosystems sustain life support systems results in even more pressure on the managers of such
systems to optimise their management effectiveness whilst still allowing for the rapid developments of
human activities.
However, we only have a limited understanding of the workings of the coastal system (including the
human component) and the effects of our management actions. Beside these uncertainties,
ecosystems are spatially and temporally quite variable and highly interconnected systems. This makes
the task of managing them a very complex undertaking. To keep on top of such a complex task, we
need broad strategic approaches to guide us. In SE-Queensland, two such strategic choices were to
use the adaptive management paradigm (AEAM, Holling 1978, Gilmour et al. 1999, Gregory et al.
2006) and to allow strong interactions and balance between resource management, science and
monitoring effort.
Figure 1-3 The adaptive management cycle is a powerful ally in managing systems that contain
many uncertainties. Graphics courtesy Keith Sainsbury.
The choice for an adaptive management approach recognises that we need to make decisions in the
presence of many uncertain factors, one of them being what the effects our management actions will
have on our natural resources. It also recognises that our management actions have two objectives:
the obvious one is to steer the managed system iteratively in the direction of the objective we have set
for it. A less visible objective is to learn from every feedback we get while iterating through the
adaptive cycle. It is these characteristics (iteration and adaptive learning) that are brought together into
a methodological approach called Management Strategy Evaluation or MSE, initially in support of
fisheries management (Smith 1993, 1994, Sainsbury et al. 2000) and also in coastal waters (McDonald
et al. 2006).
1.3. Project brief and outputs
This document reports on a ‘demonstration of concept’ project to answer the questions if and how
MSE principles could be applied to coastal zone management.
5
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
6
The project brief is:
The Healthy Waterways project aims to examine the feasibility of the application of Management
Strategy Evaluation principles to the SE-Queensland region with respect to its water resources. This
MSE would need to include riverine, estuarine and marine components. It also needs to scope how to
include economic and social factors in such MSE.
The pilot phase is expected to focus mainly on developing the assessment models for the region and
will be using a top-down approach. The operating models to support this MSE are expected to be kept
at a high level of abstraction and implemented to reflect expert opinion.
The need to connect to and reflect the regional expertise-base is of key importance for the ongoing
support and credibility of this project. This means that a fair amount of the available resources will be
used to inform and consult with the stakeholders
Project Delivery Outputs:
1. Delivery of a qualitative strategy evaluation for the SE-QLD region to the stakeholders
2. Report the successes and challenges of coastal and catchments MSE
3. If appropriate, scope a follow up project for phase II, a quantitative MSE for the SE-QLD region
The results from the pilot project against the project brief and outputs can be found in the Executive
Summary.
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
2. MANAGEMENT STRATEGY EVALUATION
The management of natural resources harbours a range of unknowns or uncertainties (Zapert et al.
1998), ranging from conceptual understanding of the system to be managed, ways to monitor and
assess the results of management actions, and the efficacy of the management actions themselves.
Where resource management acts in a multiple-use environment, there are likely to be competing or
conflicting claims on the resource (Barange 2003, Cash et al. 2006). Often, because of the
uncertainties and in some cases, unknown consequences, the exploitation of the resource is not
appropriately addressed. On the resource management level, mechanisms such as the
precautionary principle and the adaptive approach to resource management are used to deal with
some of these uncertainties.
To facilitate the generation and presentation of trade-offs for making decisions in the management of
natural resources (e.g. space, water), mathematical models are often used to simulate our
understanding of the responses of the resource to alternative management activities. As
management becomes more complex, the science-support for the decision-making process needs to
adjust to these growing challenges. Management Strategy Evaluation (MSE) is one way to tackle
the growing complexity by modelling the different parts of the management process and presenting
the manager with a set of trade-offs for each of the evaluated management scenarios whist dealing
explicitly with uncertainty.
The MSE framework was originally developed and used in fisheries management (Smith 1993, 1994,
Sainsbury et al. 2000) and is also known as Operational Management Procedures (Butterworth and
Punt 1999) The MSE approach contains conceptual elements such as management objectives,
performance measures, indicators, management scenarios and strategies. Many of its concepts are
borrowed from Adaptive Environmental Assessment and Management (AEAM, Holling 1978). This
report focuses on the role of MSE in evaluating options for the management of waterways. This
chapter examines the concepts of MSE and its defining characteristics. We also discuss the
implementation of an MSE and a key deliverable from an MSE: the management support table.
2.1. Management Scenario and Strategy Evaluation
As the pressures on the world’s ecosystems increases (Sainsbury 1991, Jackson et al. 2001, Cox
2002), the need grows for tools to help manage and conserve them. Management Strategy
Evaluation (MSE) is a set of standards and outputs that allows policies and management strategies
to be evaluated for their impacts on social, environmental and economic values. A management
strategy is a set of rules that transform the results of an assessment into management actions, given
some knowledge of the system under management. An explicit management strategy allows us (in
principle) to close the adaptive management loop by feeding the results from the assessments from
the previous iteration of the adaptive loop into the strategy rules, thus allowing us to generate the
management actions for the next iteration. By choosing contrasting management strategies, we
than can evaluate the effectiveness of different sets of management strategies (rules). We call this
process, together with a set of standards and deliverables, Management Strategy Evaluation.
However, we often do not have (access to) a set of explicitly formulated management rules and we
need to choose sets of management actions in response to assessment results. A management
action refers to a specific activity in response to an issue or event. Examples of management
actions for the waterways include upgrading a sewage treatment plant, or setting mandatory
standards for water-sensitive urban design (WSUD) for all new developments, riparian restoration of
eroded gullies or, perhaps more interesting, various combinations of such actions. A set of
management actions (and their level of implementation over time) is referred to as a management
scenario. Testing such a set of management actions (in the context of a set of given models) is
called a Management Scenario Evaluation. As the standards and the deliverables are the same for
strategy and scenario evaluation, the term MSE will be used in this document to indicate both
strategy and scenario evaluation and the particular form will be clear from the context.
MSE aims to present the trade-offs arising from various options available while managing a resource
in an adaptive fashion.
7
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
MSE delivers capability to systematically evaluate management strategies/scenarios by
presenting a set performance indicators that allows consideration of trade-offs between
those strategies/scenarios. Managing the multiple uses of resources is often a requirement
The trade-offs between various management scenarios are often expressed in measures that were
used to set (operational, measurable) management objectives. Such measures indicate how well
management is performing against the objectives and are referred to as performance measures;
they inform the manager about the discrepancy between the set objective and the actual status of
the system under management. For example, a performance measure may indicate how well a
particular water quality indicator (e.g. the total phosphorus concentration in the water) is tracking
against a reference value. Other performance measures may inform us on the effects of a given
management scenario on overall economic activity. Reporting on the effects of a given management
option in terms of performance measures often condenses a wide range of collected data into an
informative, high-level indicator of how a management option is performing. The ability to evaluate
these performance measures for a range of different management options allows us to inspect the
trade-offs between the different management options. To facilitate decision-making (choosing
between trade-offs), a systematic evaluation of each performance measure against each
management scenario is needed. Such a table of performance measures against management
scenarios is referred to as a (management) decision support table. .
The development of the Management Strategy Evaluation Framework hinges on putting the adaptive
management principle into practical use in order to further strengthen and consolidate the
management of the region’s waterways. The adaptive management approach means that
management actions are altered in response to changing circumstances. At the same time, the
approach recognises that actions can seldom be postponed until we have “enough” information to
fully understand the situation.
Operationally, adaptive means that we can change our (management) behaviour based on what
we’ve learned from our previous experience (trial, error AND learn). To be able to learn, we need to
be informed of the results of our actions. Incorporating the results of our previous actions in shaping
the decision of the next actions that should be taken is the essence of adaptive management. The
feedback loop is the essence of adaptive management. The management decisions are partly or
wholly based on the feedback of the results from the previous iteration.
The MSE approach aims to simulate adaptive management systems, so dynamic feedback is
one of the characteristics of a MSE system.
In the HW-MSE, the feedback management system consists of six elements (Figure 2-1).
Adaptive management requires the ability to monitor progress toward a target condition and to
explain this progress in terms of both responses to management actions and responses to other
changes in the situation. Behind this simple explanation lies a need for an in-depth understanding of
the nature of the changes that have occurred, and the way in which the system being managed
responds to pressures, both human-induced and natural. Because ecosystems are generally
inherently complex, and have a multitude of factors operating on them, adaptive management
requires an in-depth understanding of the way in which systems, both natural and human, operate.
As a result of this complexity it is no simple matter to identify the changes resulting from
management actions and then to clarify lessons arising from this process that will influence the
direction of management interventions. This is particularly true when operating at the scale of a
region as large and diverse as South East Queensland.
8
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 2-1 The six elements of an adaptive management system.
Observing the responses of the system under management is necessary to allow dynamic feedback
and adaptive management to occur. Many traditional modelling approaches do not include the
adaptive mechanisms in their approaches. For the purpose of the prototype HW MSE, this dynamic
feedback is initially constrained by using pre-defined sets of management actions in the SEQ
Healthy Waterways Strategy 2008 as management scenarios. In the future, though, the “Learning”
and “Management Decision” elements will be implemented to be dynamic.
In actual resource management we do not have perfect knowledge of the results of our actions. The
results from the ‘responses to management actions’ box in Figure 2-1 could be (in principle) fed back
directly to the ‘management decision’ box. However, in the real world this could be done only if we
were to have perfect information about a system’s responses to the actions. Often, we only have a
sparse subset of that information. The design of a monitoring program (spatial and temporal), which
is reflected in the observation box, is critical for the robustness of the system understanding
incorporated in the ‘responses to management actions’ box and consequently how we are tracking
the response of management actions. Simply said, the observation box acts as a filter or snapshot of
a complete system understanding.
The MSE approach aims to incorporate all major elements of ‘real world’ adaptive
management (including their restrictions) that would significantly influence the performance
measures
It is widely accepted that monitoring data have sources of error and variance. Error may be caused
by instruments not indicating the precise value of what it is measuring or observers reading an
instrument with finite precision. Variability in measurement may be attributable to the natural
variability in the underlying process that we’re trying to measure (e.g. amount of algae in the water is
affected by a number of other, non-observed, variables). For instance, if we would have sampled 10
minutes later or 10 meters away from our actual sample, chances are that the value would be
different from the actually measured value. Such process variability is not an error that we try to (or
even can) get rid of, but an intrinsic aspect of the process we are studying. The result of observation
error and process variability (by no means the only sources) will be referred to as uncertainty.
Managers often need to make decisions based on information that contains a degree of uncertainty.
The MSE approach is to explicitly deal with the uncertainty in its models.
In summary, MSE delivers capability to systematically evaluate management scenarios by
presenting a set performance indicators that allows consideration of trade-offs between those
scenarios. MSE aims to represent dynamic feedback; it includes all major elements of resource
management and deals explicitly with uncertainty.
2.2. Scoping an MSE application
MSE delivers a capability to planners, managers and scientist to evaluate scenarios for management
of our waterways. The development and application of MSE as a science-support activity has to be
seen in the broader context of the adaptive resource management process. Figure 2-2 depicts the
9
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
interactions between the actual (‘real-world’) adaptive management scheme (blue circle, see Figure
1-3 for more details) and the MSE. MSE has a strong link with the planning process as it helps
evaluate alternatives for the whole management program, including the monitoring and assessment
of the efficacies of proposed management actions or the decision-making process (management
strategies).
Figure 2-2 MSE interacts with the planning, evaluation and learning and adjustment elements
of the adaptive management cycle.
In the evaluation and learning part of one iteration of the actual management process, the learning
resulting from the evaluation is not only fed back into either a new iteration of the planning or an
adjustment of the management actions, it can also be used to update the information in the MSE. As
such, the MSE becomes a repository of the management experience and a valuable knowledge
management tool. To decide how to adjust the management actions or approaches to decision
making, the MSE can be used to trial alternatives, as we would do during the planning phase.
From the perspective of developing an actual adaptive management plan, we can recognise the
following parts from the Adaptive Environmental Assessment and Management framework (AEAM,
Holling 1978):
1. elicitation and clarification of the management objectives,
2. turn management objectives into specific, quantitative and measurable performance
measures,
3. identify a range of management options (including monitoring, assessment and
management actions/decision rules),
4. identify and quantify uncertainties in all stages of the management process,
5. predict outcomes, and,
6. communicate results to decision makers.
Management objectives are essential for an MSE. Having clearly stated management objectives is
one of the characteristics that set adaptive management apart from reactive management. The
objectives form the yardstick against which we measure success or failure. The performance
measures that are derived from the operational management objectives are used as the feedback to
assess how well the managed system is travelling in the direction of the objective set for it.
To transform high-level management objectives to operational objectives, management objectives
(Step 1) need to be analysed and turned into an operationally meaningful form. For example, an
objective cast in an abstract term, e.g. ‘protection of key habitats from degradation’ needs to be
expanded to specify what ‘key habitats’ are and what we mean by ‘degradation’. An operational
10
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
objective for the purpose of the MSE could be “the total area of seagrass meadows with a density of
at least 80% cover will not fall below the 1996 level of 3.45% of area within Moreton Bay”. In this
case the performance measure could be the difference between the actually measured total area of
seagrass in Moreton Bay with density greater than 80% and the reference value of 3.45%. The
criterion for an acceptable performance measure is whether we can measure and assess it.
Figure 2-3 An MSE simulates the essence of an adaptive resource management system.
The diagram in Figure 2-3 shows the links between a schematic representation of the elements of
actual adaptive management (front ‘slice’) and a MSE (two ‘slices’ in the back). Even though there is
no reason why any or all MSE activities could not be performed manually (and in practice they
sometimes are), we assume for the remainder of the report that the MSE functionality is
implemented in software, referred to as the MSE application. Details of the software are discussed
in Appendix B.
Apart from the feedback that is happening in the adaptive management loop and its MSE
counterparts, there are other feedback loops drawn between the management and the MSE parts in
Figure 2-3. The first feedback (solid red lines) is labelled “Inform” and “Learn”. This feedback
indicates the importance of strong interactions between the actual management activities in the real
world and the MSE system. If an MSE can be applied as a testing ground for ‘everyday’ questions
(what-if scenarios), it would be most useful to support and enhance discussion about management
actions (the inform part). If the MSE application is flexible enough to be updated according to
changes in the real world (learn part), the management organisation would have created a system to
support one of its key activities. The requirement of the MSE application to support the updating our
knowledge, together with the need to test an ever-changing collection of scenarios, dictates the need
for the MSE application to be flexible. This flexibility has consequences for the MSE design and
application software (see also Appendix B).
2.3. The MSE elements
In this section we will put a bit more details in each of the six elements (‘boxes’) of an MSE system
(Figure 2-3).
The response models represent our understanding of the response of the ‘real world’ to the
management (and its exploitation) based on our best knowledge of the system or resource. They
may include models of the ecosystem (biology, environment and their interactions with human use),
economy, water quality and quantity etc. The most challenging and time-consuming part of the MSE
11
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
is often the development of an adequate representation of the system’s responses to management.
The descriptions may be at different levels of detail, from broad-brush spatio-temporal land-use
changes down to simulating detailed effects of the application of water-sensitive urban design
(WSUD) or riverbank erosion models.
The observation models simulate the way we track or monitor the ‘real world’, typically through field
programs. Such programs may be used to extract information about the status of parts of the system
under investigation but also to assess efficacy of management. These programs are costly to
implement and maintain, and it is important to properly design the spatial and temporal
characteristics of such field programs and carefully choose the indicators they collect. For realistic
simulation of decision rules, we may want to constrain the observations from the operating models to
those that can be feasibly collected through a monitoring program. On top of monitoring the status of
the resource, additional observations may need to be collected to assess and report the efficacy of
the management itself. For SE Queensland, such monitoring schemes would include the existing
Ecosystem Health Monitoring Program (EHMP), which monitors the freshwater, estuarine and
coastal waters, and the event-based monitoring program (e.g. high-rainfall contingency monitoring).
The assessment element simulates the reporting phase and contains (often statistical) methods to
turn the collected data into management performance measures. This activity may be as simple as
drawing some summary statistics from the monitoring data or a complex as expressing an
ecosystem’s ‘health’. For SE-Queensland this would include for instance assessing the ecosystem
health index or EHI (Pantus and Dennison, 2005) and the underlying compliance values as the
precursors for the annual environmental report card.
The learning model looks at the discrepancies between the expected results of management actions
and the actual results after they have been applied to the response model. Management decisions
often include some expectation of the efficacy and effects of the management actions (controls,
levers). Learning often means simply updating those expectations. However, learning may also
include switching the overall approach for making decisions, for instance from a set of simple
heuristics (if this happens, do that) to quite complicated statistical schemes of optimising some cost
functions in the presence of uncertainty. There are two modes of learning: the passive learning aims
to update the importance of each management action after one iteration through the adaptive
management loop. The active learning mode may schedule management activities purely aimed at
gaining information about the responses of the managed system to management actions.
The management decisions element results in a set of management actions. Management actions
may entail various tools or management levers to manage (the use of) a natural resource. In the
case of coastal management, these activities may include various spatial restrictions of land-uses
(e.g. urban development), the implementation of various urban design standards, simulating the
results of behavioural change using targeted incentive schemes or whatever other management
levers are available. A resource is often managed through either controlling its exploitation (e.g.
water use restrictions) or via remedial actions (e.g. riparian revegetation). Clear, measurable
management objectives and derived performance measures are necessary to give management
decisions a goal to work towards.
The management actions element converts management decisions into actions. One of its functions
is to add implementation uncertainty: not all actions will be implemented exactly conform the
planning time lines. Another function of the management actions element is to allow the specification
of ‘fixed’ management actions. Arguably this element could be dissolved into management decision
and response models. However, this element was made explicit to address the complications of
implementing management actions.
After having put some detail against each of the elements in the MSE scheme, there is one last
issue to be discussed in this chapter: fixed versus dynamic management decisions.
To be able to simulate the decision making process in real-world management, the rules and
schemes used to make decisions need to be made explicit. In practice, decisions are being made
with often only very imperfect data available. Often decisions are made based on a mix of many
arguments, and only a subset of these arguments is based on the assessment of the feedback data.
Instead of using explicit management rules in the management decision element to transfer the
assessment results into management actions, in practice an MSE can also be used by choosing a
set of management actions, maybe varying over time, and run such a scenario of management
actions. In such cases the MSE functions as if the management decisions are fixed and do not
12
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
respond to the feedback. In doing so, this approach makes observation, assessment and learning
obsolete. Such an approach is quite common and, at best, is just a stage in the development of truly
adaptive management with explicit management rules. The ultimate goal of an MSE is to test the
effectiveness and adequacy of a complete management system, including a set of explicitly defined
management decision rules or strategies.
2.4. Other functions of MSE
Apart from its capability to evaluate management scenarios in an adaptive setting, the second
function of an MSE is to integrate the results of a range of management activities such as objective
setting, decision making, application of management actions, monitoring and reporting etc. Chapter
3 shows, in more details, how this works in the case of SE Queensland.
A third aspect of adaptive management (and MSE) is its potential for active learning. The main aim
of management is to steer resources towards achieving the management objectives. However, a
corollary, but important aim is to ensure that management is done most effectively (i.e. cost-
effective, faster benefits or outcomes, less uncertainty, etc.) This corollary aim is an essential
component of ‘active learning’. To ensure that active learning process can occur, there must be a
willingness to develop and trial management actions with a primary aim to learn more of their
effectiveness. In most cases, this activity is under-appreciated especially in the common situation of
over-stretched management resources or risk-averse management practices. Such active learning
objectives, explicit management objectives and management decision rules set adaptive
management apart from reactive management.
The fourth function of an MSE is that it allows the experiences from management to be consolidated
into the system: MSE can also function as a knowledge management tool.
It is not a task of MSE to assess the correctness of management objectives. However, it allows us to
assess their attainability in terms of their performance measures and based on our best knowledge.
2.5. Summary
This chapter gave MSE a place in a real-world adaptive management scheme and proposed and
discussed six elements of our model of resource management. Emphasis was placed on the
adaptive character of MSE through management decision procedure or rules, also called a
management strategy. We touched on the management scenario (manual-decisions) form of MSE
as a transition phase and concluded that clear, measurable management objectives are needed to
direct our actual management, as they do the decisions made within a MSE. Apart from evaluating
management scenarios, an MSE can be used as an integration platform; it promotes active learning
and can function as a knowledge management tool.
13
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
14
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
3. CATCHMENT-TO-COAST MSE IN SE-QUEENSLAND
In this chapter we will examine some SE-Queensland water management issues and how MSE can
be used to support the management of these issues. This chapter also introduces the Logan-Albert
MSE demonstration project.
3.1. SE-Queensland water management issues
There are a range of water-related issues mentioned in the Strategy. These include the current and
future pressures on waterways in SE-Queensland, often caused by and causing social and
economic issues. When asking experts to ‘associate freely’ on the subject of regional water
management, they came up with a wide range of related issues, summarised in Figure 3-1. We
expect that many of these issues would be common for densely populated coastal areas within and
outside Australia.
Figure 3-1 The collective mind-map of experts in SE-Queensland regarding the region’s water
issues.
The issues range from social (e.g. indigenous, community, demographics), economic (e.g. cost of
potable water, developments) through to ecological (e.g. ecosystem health, habitat pressures).
Water issues touch almost every aspect of our society and managing water is a crucial task,
especially in area where the use of water threatens to exceed its availability.
Discuss recent changes to people’s attitudes to water issues, climate variability, water security and
infrastructure.
To put SE-Queensland in perspective: Queensland’s land area is over 1,735,000 km2 which is about
0.3% of the Australian mainland area and about 22% of the Australian mainland area. SE-
Queensland land area is around 23,000 km2 (Table 3-1) or about 1.3% of Queensland’s area.
Queensland’s total population is little over 4 million people of which 2.7 million or 67.6% live in SE-
Queensland. The combination of rapidly increasing population, drought, and a strained water
infrastructure as can be found in SE-Queensland, highlights the importance of water management.
15
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 3-2 SE-Queensland has a range of water-related management issues (graphics
courtesy of the SEQ-HWP).
Figure 3-2 illustrates some of the many coastal issues and their relationship with the main of
Moreton Bay catchments. The biophysical and ecosystem health issues are mainly arranged around
water quantity and three constituents: nitrogen, phosphate and sediment. The sources of these
constituents are arranged in diffuse loads (sediment or nutrients from catchment runoff, groundwater
inputs or atmospheric fall-out) and point-sources (single point of pollutant discharge such as sewage
treatment plant or an industrial wastewater treatment).
3.2. Applying an MSE approach to SE-Queensland water issues
There are obvious and strong links between developing a MSE and the resource management
strategies set for a region. The Strategy 2007 – 2012 document identifies a range of measures to
improve the understanding, protection and conservation of our catchments and waterways, while
maintaining our lifestyle and livelihoods’. The Strategy indentifies the following environmental
threats to our waterways:
urban diffuse source pollutants (e.g. nutrients, sediment and litter)
physical disturbances to urban waterways resulting from changes in stormwater flow rates,
volumes and flow frequencies
point source pollution (e.g. stormwater and sewage treatment plant outfalls)
pressures on areas and waterways of high ecological value
algal blooms in estuarine and coastal marine waters, including Moreton Bay
deteriorating catchments such as Lockyer, Logan-Albert, Bremer
These threats will guide future MSE developments, especially the choice of the response models
that need to be developed. For instance, we need to simulate the response of diffuse constituents
(nutrients, sediment) coming from our catchments to various management actions such as regulating
urban development or riparian revegetation if we want to look at management options to manage
diffuse runoffs.
W
Wa
at
te
er
r
q
qu
ua
al
li
it
ty
y
I
In
n
s
st
to
or
ra
ag
ge
es
s
U
Ur
rb
ba
an
n
s
st
to
or
rm
m
w
wa
at
te
er
r
S
Se
ea
ag
gr
ra
as
ss
s
l
lo
os
ss
s
t
th
hr
ro
ou
ug
gh
h
L
Ly
yn
ng
gb
by
ya
a
P
Pa
as
st
tu
ur
ra
al
l
u
us
se
e:
:
r
ri
iv
ve
er
rb
ba
an
nk
k
e
er
ro
os
si
io
on
n
A
Aq
qu
ua
ac
cu
ul
lt
tu
ur
re
e
r
ru
un
no
of
ff
f
S
Se
ew
wa
a
g
ge
e
D
Di
if
ff
fu
us
se
e
a
ag
gr
ri
ic
cu
ul
lt
tu
ur
re
e
r
ru
un
no
of
ff
f
I
In
nd
du
us
st
tr
ri
ia
al
l
w
wa
at
te
er
r
u
us
se
e
P
Pl
la
an
nt
ta
at
ti
io
on
n
f
fo
or
re
es
st
ts
s
c
cl
le
ea
ar
ri
in
ng
g
16
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
In Chapter 2 we discussed the needs for clear and measurable management objectives to guide the
MSE’s decision-making element. The Strategy also states a range of targets and measures to
achieve them. Some of these are particularly important to the development of an MSE (objectives
are underlined, actions are in italic):
major reductions in urban diffuse source pollutants through adoption by 2008 of Water
Sensitive Urban Design (WSUD) objectives for all new developments;
major reductions in urban diffuse source pollutants by putting strategies in place by 2011
(including staged interim retrofit targets) for meeting local, catchment-specific performance
standards for WSUD in 100 percent of existing urban areas;
reduction in loads from point source pollution by reuse of treated water (up to 100% of
average dry weather flow where demand exists for beneficial reuse);
reduction in concentrations of key pollutants (down to 3 mg/L nitrogen and 1 mg/L
phosphorus) where beneficial reuse is not possible;
significant reductions in non-urban diffuse pollutants (particularly sediment) with a target
of a 50 percent reduction in sediment load to Moreton Bay by 2026; and
support for a range of comprehensive monitoring programs and the further development
of predictive modelling tools.
Some of the expression of management objectives may need further examination to make them
directly applicable within a MSE framework. For instance, ‘major reductions’ is not a measurable
objective and needs to be replaced by an objective stated along the lines of ‘a reduction of X
percent’. The Strategy goes on and lists about 500 specific management actions to be implemented
over the 2007-2012 period.
The Strategy also mentions ‘further development of the Management Strategy Evaluation framework
for waterways in South East Queensland, including improved capacity to model adaptive
management scenarios as a measure it has adopted to reach its vision’.
3.3. MSE as a platform for integration
There are a range of activities (Figure 3-4) currently in place in SEQ that would be in need of
integration.
We mentioned in section 2.4 that MSE can be used as a platform for integration. This section goes
into some detail to expand that point with respect to the SEQ-MSE.
Figure 3-3 shows a selection of activities that are undertaken in SE-Queensland that feed into the
water resource management. Most of these activities are producing results, often independent of
other activities. For instance, the catchment models (EMSS/E2) are delivering results about the
flows and constituent runoffs (sediment, nutrients) of a catchment. WSUD models (such as MUSIC)
deliver results on stormwater management options, but are separate from the catchment models. By
using separate models, the potential exists that the results from these models were run with
different, or even conflicting, assumptions. The results of these models thus need to be integrated
by the receivers of this information, often the resource managers themselves.
17
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 3-3 A range of activities and information is used in the management of SE-Queensland
waterways.
Alternatively, integrating such functionalities will help us better inform resource managers of the
effects of various management actions. For instance, bringing together both models in an MSE
application (the catchment model as part of the response element and the stormwater model as part
of the management actions or management decision element) and thereby integrating their working,
gives them a recognisable place in the management scheme of activities.
To examine the notion of MSE as an integration platform, Figure 3-4 shows some of the key
activities of the HW Partnership on the left-hand side On the right hand side of Figure 3-4, the two
‘slices’ represent the MSE framework.
18
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 3-4 MSE as an integration platform: many activities (and information) find a place in
the MSE framework. The MSE framework effectively connects and thereby integrates these
activities.
The six coloured boxes in Figure 3-4 represent the now familiar elements of the MSE approach. The
coloured, horizontal arrows indicate where various activities or information from the Partnership may
be used in the MSE process. The double-sided arrows indicate that an MSE not only uses
information from the management process, but that it may also change those activities in turn. For
instance, a logical first step is to configure the observation element in the MSE application in the way
the existing monitoring program is conducted. However, once the MSE system is fully operational,
we can examine the consequences of changes to the monitoring program (e.g. changes to its spatio-
temporal design) with respect to the overall outcomes of the management or changes of the
management decisions. A second example would be to trial the planned on-ground actions as
defined in the Strategy 2007-2012 as scenarios using the MSE approach. The proposed actions can
be represented as management levers for the management decision in the MSE system. The rollout
of proposed management actions could also be tested as pre-defined (non-adaptive) sets of
management actions. The results of these MSE scenarios may feed back into the Strategy in the
form of changes in (combinations of) management actions or their timing.
From a science point of view, the learning and decision making processes are of key interest as
simulating decision making in such a complex and uncertain environment poses a considerable
challenge. Looking at the efficacy of governance under different institutional arrangements would be
an area of interest of social sciences that would fit in well with a MSE framework by simulating them
as part of the management decision element.
A range of projects regarding the SE-Queensland economic and social management levers and
response models, industry activities and management options costing etc. would fit on the SE-
Queensland science development agenda.
The ecosystem services approach is deemed to have attractive properties (economic basis,
balancing human utility and ecosystem integrity) to be part of a MSE. Further examination of how to
apply the potential of ecosystem services in a C2C MSE would also be part of a future science work
plan.
19
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
3.4. The Logan-Albert demonstration project
One of the action plans in the Strategy 2007-2012 is called the Management Strategy Evaluation
Action Plan. This Action Plan has at its core a trial of the MSE approach to the complex environment
of SE-Queensland water management. It aims to integrate freshwater, estuarine and marine
ecosystems and the multiple management regimes operating in the region, as well as the human
and natural impacts on these systems. To demonstrate the concepts and workings of a C2C-MSE, a
one-year pilot project was commissioned. The description of the projects and the report on its
findings are the subject of the remainder of this document.
The scoping part of the pilot study consisted of making some choices regarding what to include in
the HW-MSE application. To include the whole of the SE-Queensland region in the pilot study would
detract from its main objective: to demonstrate what a Catchment-to-Coast would look like.
Consequently, the first task of the pilot project was to choose a SE-Queensland area for its
application.
Catchments are relatively well defined units with respect to water management, and the first choice
was to use a catchment as the area of interest. The subsequent choice to be made was which
catchment to use. Initially two catchments were proposed: the Brisbane and the Logan-Albert
catchments, the two biggest catchments in SE-Queensland (see Table 3-1). The choice for these
two candidates was based on availability of data, the level and nature of the pressures on the
catchments (present and future) and the feasibility of the demonstration project (complexity of
issues).
The criteria used to decide between the two catchments for the MSE pilot study were the range of
issues to be addressed and the availability of information and its ecosystem health status. The
choice for the Logan-Albert (the second-biggest catchment, see Table 3-1) was based on its
expected rapid population expansion, diversity of issues and availability of a calibrated catchment
model (hydro-dynamics and transport). The Logan-Albert was also explicitly mentioned in the
Strategy 2007–2012 document as one of the ‘deteriorating catchments’.
Table 3-1 The projected mainland areas of the SE-Queensland catchments.
Catchment name Area [km2]
Upper Brisbane 5,435
Logan/Albert 3,859
Lockyer 2,959
Bremer 2,025
Stanley/Somerset 1,518
Nerang/Gold Coast 1,317
Pumicestone 1,244
Lower Brisbane 1,188
Noosa 848
Pine Rivers 819
Maroochy 627
Mid Brisbane 548
Redlands 279
Mooloolah 221
Total SE-Queensland 22,887
After the catchment was selected, the pilot project’s first task was to scope the overall design of the
Logan-Albert MSE application. The result is shown in Figure 3-5.
20
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 3-5 Sketching the MSE as a conceptual diagram is a useful first step in designing a
MSE application. We used IAN (http://ian.umces.edu/symbols) symbols to construct this
graphic.
Figure 3-5 shows the elements of the adaptive management system and as such forms the template
for the MSE application. The six boxes make up the elements in a MSE system (see also Figure
2-1). The red text indicates the parts of the adaptive management framework that are already in
place in the region and can be drawn upon when populating the pilot MSE application.
The response model in Figure 3-5 describes our knowledge of the system, especially where relevant
in response to management actions. For the pilot project, this element included a catchment model
that transforms rain into water flows and transports constituents such as sediments and nutrients. It
also contains a simple aquaculture economics model to demonstrate the capability of MSE to
integrate over a range of sectors (e.g. biophysics, economics). The response model is also exposed
to external drivers such as climate and demographic changes.
The observation model Figure 3-5 allows us to simulate the current estuarine and marine EHMP-
sampling regime with monthly samples and any other sampling regime we may be interested in. In
Chapter 5 we will demonstrate the use of various sampling regimes in examining the efficiency of the
monitoring program.
The assessment model Figure 3-5 is able to generate various performance measures, the report
card grades may be one of these measures. Other performance measures may report on the effects
of management actions on economic or social processes.
The Learning and Decision models are of special interest as there are currently no explicitly defined
processes in the region to deal with them. We expect over time that they will be developed in
collaboration with the stakeholders and the science providers in the region.
The management actions model would implement a range of management levers that allow us to
apply various management actions from the Strategy 2007-2012 to the response model.
The next stage is to design a MSE application based on the conceptual diagram in Figure 3-5. The
experience with developing fisheries MSE applications is that they rapidly grow to a high level of
complexity, partly due to the number of models involved and the need for communication between
them. The need for a flexible approach regarding the ability to ‘plug and play’ different models and
to deal with uncertainty puts a fairly heavy burden on the software design and implemented
21
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
infrastructure. However, if these fundaments are not in place, demonstrating the functionality of a
MSE application becomes very hard.
As discussed in Chapter 2, MSE is a complex and often abstract subject, especially for those who
are new to the field. One of the critical success factors for the acceptance of MSE by resource
managers is how well its capabilities can be demonstrated on realistic questions that may arise in
practice. We found that good support for the workflow arising from MSE (creating models,
composing scenarios, evaluating and post-processing scenarios) is needed to demonstrate the
considerable potential of MSE to the prospective users of such an approach. An informative and
systematically arranged user interface is part of the demonstration software application and is a key
deliverable from the pilot project.
A MSE often uses a range of models and a particular scenario is constructed by combining different
models. The flexible use of models is another important function that the software application needs
to fulfil. The software development task of the pilot project spend considerable part of the available
effort in developing a novel communication mechanism, standards for ‘plug & play’ functionality that
allow the flexible use of different models.
For the pilot project, we selected two models that would be able to demonstrate the ways of
composing a response model out of sub-models. We also selected the assessment methods based
on the ecosystem health and compliance already in use in SE-Queensland and two management
levers, land-use based management actions and remedial management actions.
To better understand the Logan-Albert catchment, the next section gives a broad overview of its
characteristics.
3.5. The Logan-Albert catchment
The HW-MSE pilot project has chosen the Logan-Albert catchment as its main focus. This section is
intended to give a broad-brush sketch of the region. Figure 3-6 shows the geography of the Logan-
Albert catchment and its location amongst the other catchments in SE-Queensland.
22
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 3-6 The Logan-Albert catchment is one of Queensland's most southern catchment. Its
main rivers are the Logan and the Albert (southern branch on the image).
The Logan-Albert catchment is the second-largest catchment in SE-Queensland (see Table 3-1) with
a projected area of around 3,850 km2. Logan-Albert catchment borders the Moreton Bay and
contains one of the four major rivers (together with the Brisbane, Pine and Maroochy) that flow into
the Moreton Bay. The SE-Queensland population is projected to grow from 2.6 million to 3.7 million
between 2004 and 2026 (see Table 3-2). The population of the Logan-Albert catchment is expected
to grow from 280,000 in 2004 to 380,000 in 2026, a growth of over 35%. The numbers of the
population growth are based on local government area (LGA) data supplied by the Queensland
Office of Urban Planning. The data in Table 3-2 was derived from a spatial overlay of the original
planning projections per LGA with the catchment boundaries. The assumption for the resulting
numbers in Table 3-2 is that the population is homogeneously distributed over each LGA.
23
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Table 3-2 Population growth predictions (x10,000) (original data: Qld Office of Urban
Planning)
Catchment 2001 2002 2004 2016 2026
Lower Brisbane 91 93 96 110 130
Nerang/Gold Coast 38 39 43 55 64
Logan-Albert 26 27 28 34 38
Pine Rivers 19 20 20 25 28
Redlands 15 16 16 19 19
Pumicestone 13 14 16 22 26
Maroochy 9.1 9.3 10 14 16
Bremer 8.5 8.6 9.0 12 16
Mooloolah 7.2 7.5 8.1 11 13
Lockyer 5.0 5.1 5.2 6.2 7.3
Noosa 3.6 3.8 4.0 4.6 4.7
Stanley/Somerset 1.8 1.8 1.9 2.5 2.9
Mid Brisbane Catchment 1.1 1.2 1.2 1.6 2.5
Upper Brisbane 0.66 0.67 0.67 0.74 0.80
SE-Queensland total 239 247 260 318 368
Another important descriptor of the catchment is the amount of rainfall (the main source of water) the
catchment receives. The Logan-Albert catchment receives between 500 and 1,500 mm of rain per
year, with a median of 1,000 mm. Assuming a median annual rainfall of 1,000 mm, the catchment
captures ~3,800 GL/year (no losses). Of this volume, only 10% is discharged by the Logan/Albert
river system; the major losses are from evapo-transpiration, resulting in a reported long-term annual
flow volume of about 380 GL.
3.5.1. Logan/Albert overall statistics
Long term (1983 – 2003) averages reported for the Logan/Albert catchment by the current E2
catchment model are: annual flows~ 480 GL/yr, total nitrogen (TN) loads around 600 t/yr, total
phosphate (TP) loads of about 70 t/yr, and total suspended solids (TSS) loads in the order of 4,000
t/yr. Table 3-3 shows the areas of land-use in the Logan/Albert catchment as used in the catchment
model.
Table 3-3 Areas of land-use in the Logan/Albert catchment model
Land use Area [km2] % area
Dense urban 30 1%
Suburban 135 4%
Rural residential 205 6%
Broadscale agriculture 158 4%
Intensive agriculture 22 1%
Grazing 1439 39%
Greenspace 1700 46%
Water 4 0%
Total 3694 100%
24
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
3.6. Summary
In this chapter we examined some of the challenges of water resource management in SE-
Queensland. We also discussed the Healthy Waterways Strategy 2007 – 2012 as a response to
these challenges and to connect parts of the Strategy to the MSE we propose to develop for SE-
Queensland over time.
This chapter examined the potential of MSE to function as a platform to integrate a range of activities
in the coastal zone and show that some of the current activities are instrumental to the development
of an MSE.
We reported on the choice for the Logan-Albert catchment as the focus area for the Healthy
Waterways MSE demonstration project and came up with key success factors for the software
implementing the MSE application. We gave some ‘vital statistics’ that describe the catchment and
placed it in the context of population growth predictions.
We started with the task of deciding what models to put into the pilot-MSE application. Chapter 4 will
continue this by describing in more detail each of the models in the pilot-MSE application.
25
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
26
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
4. LOGAN/ALBERT PILOT MSE MODELS
The pilot MSE application in the Logan Albert Catchment is aimed to demonstrate the potential and
functionality of the MSE approach, as a precursor to a fully functional MSE implementation for South
East Queensland. The MSE aims to integrate the functionality of the existing models. The MSE pilot
project does not aim to develop detailed response models or obtain information to calibrate and
operate such models.
In the case of the Logan-Albert pilot MSE application, most of the response models were already
developed by the SE-Queensland Healthy Waterways Partnership. The structure, functionality and
form of the models needed to be fitted within the pilot HW-MSE application.
In the previous chapter, section 3.4, we indicated the models to be used in the various MSE
elements: (i) a catchment model, (ii) an industries model to function as components to construct a
response model, (iii) an observation model, (iv) a model to perform the assessments of the system
under consideration (as represented by the response model) along the lines of the SE-Queensland
report card values and compliance indices, iv) a model to generate land-use management actions
and (vi) a model to simulate a range of remedial actions. Each of these models will be discussed in
the next sections.
4.1. The catchment model
The catchment model implemented in the pilot HW-MSE system has two functions: conversion of
rainfall into runoff and estimating the transport of constituents such as sediment and nutrients
associated with the run-off.
The first function is performed by the rainfall-runoff model referred to as ‘SIMHYD’. SIMHYD is
derived from the MODHYDROLOG model (Chiew and McMahon (1991) (Figure 4-1). Details of the
SIMHYD model are given in the CRC for Catchment Hydrology’s report RLL, Rainfall Runoff Library
(Podger 2004). The catchment model takes into account the two main channels for water runoff:
surface and ground water; and as well as evapo-transpiration and storages.
27
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 4-1 A conceptual diagram representing the daily rain runoff model ‘SIMHYD’ (Podger,
2004).
The Logan-Albert sub-catchments used in the pilot HW-MSE application were generated using the
E2 catchment modelling package and a 50 m digital elevation model (DEM) of the catchment.
Figure 4-2 show a map of the resulting sub-catchments. The sub-catchments are represented as a
linked node network (or directed graph). Each node represents an accumulation point for the flows
(and constituents) coming from the catchments feeding into it.
Depending on the amount of rain (mm per day) and the areas of the various land-cover or land-uses
(called ‘functional units’), the SIMHYD model calculates surface and base flows, as indicated in
Figure 4-1.
28
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 4-2 The Logan-Albert E2 catchment model uses 37 sub-catchments to calculate water
and constituents runoff (coloured polygons). The 37 sub-catchments are represented by a
directed node network (the arrows and black dots).
Logan-Albert catchment is described as having eight different land-uses (functional units): dense-
urban, sub-urban, rural residential, broad scale agriculture, intensive agriculture, grazing, water and
green space. As each functional unit is described by a separate SIMHYD model (defined by a set of
seven parameters plus its area, pervious fraction and impervious threshold), there are a total of
2,960 parameters (37 sub-catchments, eight land-uses per sub-catchment, ten parameters per land-
use) to specify the rainfall runoff model for the Logan-Albert.
Apart from turning rain into water flows and water vapour, the second function of the catchment
model is to simulate the transport of constituents from the catchment, referred to as the transport
model. In the Logan-Albert catchment model, the constituents are total suspended solids (TSS),
total nitrogen (TN) and total phosphate (TP). The transport model differentiates between the surface
flows and the base flows to calculate the constituent loads for each of the three constituents. The
transport model is simply scaling the flows by two constants for each constituent. One parameter for
each constituent for the surface flow, called the Event Mean Concentration (EMC), and one for each
constituent for the base flow, called the Dry weather Mean Concentration (DMC). In total, there are
six transport parameters for each functional unit, bringing the total of 16 parameters per functional
unit and resulting in 4,736 parameters for the catchment model. The actual values of the parameters
have been estimated using field and literature data, a process known as model calibration. In
practice, many of these parameters have the same value in the current model implementation.
There are currently two external forcing series used to drive the Logan-Albert catchment model: the
potential evapo-transpiration (PET) and the daily rainfall time series. PET is a measure of the
potential for water to evaporate or be transpired in different places in Australia. Within a relatively
small area such as the Logan-Albert, we can assume PET to be spatially invariant. PET values are
available through a Bureau of Meteorology service called SILO
(http://www.bom.gov.au/silo/GenInfo.shtml).
29
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
2
2.5
3
3.5
4
4.5
5
5.5
6
6.5
1357911
Month
PET [mm]
Figure 4-3 Potential Evapo-Transpiration or PET values for the Logan-Albert region. The bars
indicate the standard deviation of inter-annual variation of PET values between 1983 and
2003.
The seasonality of PET (Zhang et al. 2001) is important and can change between years. Figure 4-3
shows a plot of the average PET values for the period between 1983 and 2003, derived from the
SILO data set. The inter-annual variation of PET is indicated by the standard-deviation bars in Figure
4-3.
30
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Avg
500
700
900
1100
1300
1500
1700
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
Year
Avg annual rainfal [mm]l
Figure 4-4 The (SILO) daily rainfall data is accumulated into annual averages for the Logan-
Albert catchment. The spatial variability over the catchment is reflected in the standard
deviation bars in the graph.
The pilot HW-MSE catchment model uses the time series from 1983 to 2003. A summary of rainfall
in the catchment is given in Figure 4-4 for the Logan-Albert region. In future implementations of the
HW-MSE, there will be more options to either randomise the years of the rainfall time series within
the stochastic replicates or to use a rainfall-generating model that is calibrated by the existing daily
rainfall series. The latter option would allow for extensive climate-change scenarios to be evaluated.
Finally, the catchment model uses the 1999 land-cover map produced under the Statewide
Landcover and Trees Study (SLATS) program, shown in
Figure 4-5. Logan Albert Catchment is comprised of 45% green spaces (natural bush and
conservation), about 40% of its area is in use for grazing and agriculture and about 10% urban land-
use (not all of the catchment has been mapped into land-uses).
31
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 4-5 The pilot HW-MSE catchment model is based on the 1999 land-use map, produced
by the State-wide Land-cover and Trees Study (SLATS) program
A sample result of rainfall and run-off (flow) is presented in Figure 4-7. These results were obtained
by running the catchment model over 20 years (1983-2003) and extracting the information for the
node at the mouth of the Logan River.
0
200
400
600
800
1000
1200
1400
1600
1980 1985 1990 1995 2000 2005
Yea r
River mouth
avg rain [mm/yr] Flow [GL/yr]
Figure 4-6 The Logan-Albert catchment model’s simulated flow, based on the SILO
daily rainfall time series.
To get some indication as to how good the catchment model is performing with respect to the actual
river flows, the results of flow simulation in Figure 4-6 were compared with measured results in Table
4-1 (as cited in Australian Natural Resources Atlas
(http://www.anra.gov.au/topics/water/availability/qld/index.html#sw_vol). The long-term natural
outflow for the Logan River is around 390 GL per year indicated in Table 4-1. The 20 year average
32
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
33
flow of the E2 catchment model is around 490 GL per year. The catchment model does not include
any diversion of water apart from the storages. Note: This apparent discrepancy needs to be further
examined and addressed before the HW-MSE can be fully applied to real management questions.
Table 4-1 The measured long term runoffs, cited from the Australian Natural Resources Atlas.
River Natural Outflow [GL/yr]
Annual Mean Run-off
[GL/yr]
Brisbane River 1,072 1,113
Maroochy 659 674
Logan 351 389
Pine 309 380
Results of the model runs of the various constituents (after simulating Logan flows) are shown in
Figure 4-7. The top row of graphs show the loads of the three catchment model constituents, total
nitrogen, total phosphate and total suspended solids. The bottom row of Figure 4-7 shows the
relative contribution of each of the land-uses and highlights the importance of grazing as a
contributor, especially to suspended solids. This relatively high contribution of grazing to the overall
catchment load is partly due to the large proportion of the catchment used for grazing (35%). The
contributions of green space to the total nitrogen load and the importance of suburban land-use to
total phosphate and suspended solids is also prominent. To enable us to perform a level of ‘intuitive’
verification of the results MSE scenario results, a broad understanding of a catchment’s
characteristics is helpful.
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
4.2. The industry models
The catchment model plays a major role in the pilot because it generates the key factors (flow and
constituents) that form the base of the existing monitoring and assessment programs for water-
related issues in SE-Queensland. To a certain extent, this reflects the main focus of research,
monitoring and assessment over the last 10 years. It is also widely recognised that these may not be
the appropriate information base for resource management scenario evaluation in the future. To
open up the MSE to include other areas (e.g. economics, social, institutional arrangements), we
included an economics model in the HW-MSE pilot application that represents the aquaculture
industry in the mouth of the Logan River and the sugar cane farming industry. However, at this stage
of the development, there are no SEQ-wide industry models or catchment relevant industry models
to integrate into the MSE. Additionally, there is very little catchment based research on industries
form, function or relations with waterways. The work presented here was also undertaken to show
what direction such research may take in the future.
4.2.1. Aquaculture
The Logan Albert aquaculture industry is one of the few industry’s with a direct and pertinent link
between economic production and waterway quality. The model derived for the MSE was developed
during the pilot phase and based on various source (eg. Brennan, 1999; Lobegeiger 2008). However
the industry is quite small and many characteristics of the industry are commercial-in-confidence, as
such down scaling from SEQ known data was necessary to estimate the magnitude and direction of
relationship between water quality and turnover of the industry.
The essential elements of the aquaculture farm economic model are outlined in Table 4-2. For every
7 ha, average farm size, an estimated 1.1 full time equivalent (FTE) jobs were created. The
economic model looked at alternate enterprise structures (i.e. open, flow-through and recirculation)
which depends heavily on the water quality that enters and leaves the farm and outlines the
production outputs (tonnes per pond ha), gross income and variable cost ($/pond ha) and hence, the
farm operating return for the ‘average’ farming practice.
35
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Table 4-2 The basis of the aquaculture industry model is a model simulating the economics of
a aquaculture farm in the Logan-Albert region.
Production Option Aquaculture
Farm Size Ha 7.00 9.00 9.00
Enterprise Open Flow-through Recirculation
Area Harvested 5.00 5.00 5.00
Area required 7.00 9.00 9.00
Production per ha (t/pond ha) 5.40 5.40 5.40
Gross Income ($/pond ha) 19,878 19,878 19,878
Variable Cost ($/pond ha) 5,576 7,576 7,576
Gross Margin ($/pond ha) 14,302 12,302 12,302
Area Gross Margin ($/ha) 10,216 6,834 6,834
Farm Gross Income 99,390 99,390 99,390
Less Variable Costs 27,880 37,880 37,880
Farm Gross Margin 71,510 61,510 61,510
Less Enterprise Fixed Costs 40,120 41,055 41,525
Farm Operating Return 31,390 20,455 19,985
Source: Brennan 1999
The actual area under production and the production tonnage was updated to reflect the Logan-
Albert catchment conditions as found in Figure 4-8.
Logan River Prawn Sector
0
100
200
300
400
500
600
700
800
900
90/91
91/92
92/93
93/94
94/95
95/96
96/97
97/98
98/99
99/00
00/01
01/02
02/03
03/04
04/05
05/06
06/07
tonnes
0
20
40
60
80
100
120
140
hectares
Production (t) Ponded Area (ha)
Figure 4-8 The development of aquaculture in the Logan-Albert catchment. From: Report to
Farmers - Aquaculture production survey, Queensland 1990-91 to 2006-07 (QDPI, 2008)
36
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Finally there is a need when undertaking a direct assessment of the economic impact of any industry
to put this in the context of the broader region and economy. Here we scale up our direct results
from aquaculture in the catchment to SEQ economy.
For every one million dollars of output (i.e. farm gross income) an additional $700,000 will be
generated in SEQ (Brisbane – Moreton region) in direct economic activity (where direct
economic activity is equivalent to farm gross margins). This is a type 1 multiplier for other
agriculture production for 1996-97 year (OGS, 2000).
For every one million dollars of output (i.e. farm gross income) an additional $1,000,000 will
be generated in SEQ (Brisbane – Moreton region) in direct and indirect (i.e. changes to
household consumption from change in income) economic activity. This is a type 2 multiplier
for other agriculture production for 1996-97 year (OGS, 2000)
For every one million dollars of output (i.e. farm gross income) an additional 14.5 jobs will
be generated in SEQ (Brisbane – Moreton region) in direct economic activity (where direct
economic activity is equivalent to farm gross margins). This is a type 1 multiplier for other
agriculture production for 1996-97 year (OGS, 2000).
For every one million dollars of output (i.e. farm gross income) an additional 19.1 jobs will
be generated in SEQ (Brisbane – Moreton region) in direct and indirect economic activity
(i.e. changes to household consumption from change in income). This is a type 2 multiplier
for other agriculture production for 1996-97 year (OGS, 2000).
4.3. The observation model
The observation model is by far the simplest of the models used. Currently it only allows us to
choose how often a sample is taken. Depending on which combination of response models are
used (catchment, industry), the observer requests a sample of all available indicators. The
catchment model is sampled in the river mouth for rain, flow, total nitrogen, total phosphate and total
suspended solids. The aquaculture industry model is sampled for their contributions to nitrogen,
phosphate and suspended solids. The aquaculture industry model also provides the two economics
indicators: gross farm income and variable costs, both scaled up for the region.
4.4. The assessment model
The assessment models for the biophysics responses in SE-Queensland are based on the
Ecosystem Health Index (Pantus and Dennison 2005). The Ecosystem Health Index (EHI) is based
on the compliance of regions to the various reference values set for a range of variables such as
sediment, algal chlorophyll, seagrass, nutrients etc.
The catchment model implemented in the HW-MSE application simulates only flow, sediment and
nutrient loads. Consequently, the EHI performance measure in the HW-MSE application is based on
these three variables. Table 4-3 shows the statistic and reference values used to calculate the EHI.
Table 4-3 The reference values used in the SEQRMS assessment model at the Logan/Albert
river mouth
Reporting
region Ecosystem
type Indicator Statistic
Steepness
compliance
function Operant Reference Unit
Logan River Estuary Total nitrogen
(TN) median 200 < 32.14(1.02) µM (mg/L)
Logan River Estuary
Total
phosphorous
(TP) median 200 < 1.875 (0.026) µM (mg/L)
Logan River Estuary Turbidity median 200 < 20 NTU
37
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
The assessment of the aquaculture response model is based on some characterising statistics of the
economics such as annual gross margin from the catchment-scale aquaculture industry, and its
effect on SE-Queensland economy and employment.
4.5. Institutional arrangements
One of the topics of discussion is the importance of the institutions (and their relationships) involved
in managing the coastal zone’s water resources and how to measure the sometimes intangible
institutional outcomes of water management partnerships
The Moreton Bay Waterways and Catchment Partnership is a whole of government / whole of
community strategic framework for the integrated and sustainable management of the waterways
and catchments of the South East Queensland region, including Moreton Bay and coastal zones.
The philosophy underlying the Partnership’s approach rests on two foundations: commitment to
working in a coordinated partnership structure in which all partners can be heard, contribute to
decision-making and implement agreed actions within their own spheres of responsibility; and the
formulation of management strategies on the basis of sound science, rigorous monitoring of the
waterways environment, and adaptive learning (SEQHWP, 2007).
Its vision is that “By 2020 South East Queensland’s waterways and catchments will be healthy
ecosystems supporting the livelihoods and lifestyles of people on south-east Queensland, and will be
managed through collaboration between community, government and industry.” To achieve this
vision, thirteen objectives have been identified which form the structure of how the vision is
achieved.
1. Protect the High Ecological Value waterways of South East Queensland.
2. Reinstate the ecosystem health of disturbed and degraded waterways.
3. Maintain the Moreton Bay environment and its biodiversity, including the survival of
significant species (including dugong, turtles and fish) and biological communities (including
wetlands and seagrass).
4. Implement effective total water cycle management practices supporting ecologically
sustainable development and the community’s environmental values.
5. Ensure water quality appropriate for: sustainable industry and agriculture, recreational
amenity, including safe swimming and secondary contact activities such as boating, visual
amenity, and indigenous use.
6. Build appreciation of, and motivation to maintain, the cultural, historical, amenity and
economic values of the region’s waterways.
7. Maintain, and where necessary improve, the health of catchments and waterways to ensure
their contribution to the long-term security of water supplies in the region.
8. Maintain and enhance effective institutional arrangements at all levels.
9. Invest in continually improving the knowledge, information base and tools to facilitate
efficient and effective management.
10. Conduct activities and programs in an efficient, professional, transparent and accountable
fashion.
11. Encourage and support active collaboration between community, government and industry
to achieve the Healthy Waterways Vision.
12. Build community understanding and ownership of problems and support in achieving
effective solutions.
13. Engage the community in identifying issues and developing and implementing appropriate
solutions.
The challenge for HWP and many other natural resource based partnerships (see Robinsom, et. al.
forthcoming) is that attainment or progress along objectives is a critical success criterion for any
partnership. However, the only way to know if progress is being achieved is through measurement of
38
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
some form. For HWP the measurement of its environmentally based objectives is being achieved
through the Environmental Health Monitoring Project (EHMP). Broadly, objectives 1,2,3, 5 and 7 are
measured through the EHMP.
Additionally the partnership undertakes structured and rigorous financial assessment each year
which could address objective 8. The majority of the objectives (4, 6, 9, 10, 11, 12, and 13) could not
be measured by either the environmental or financial indicators. We focus on measuring those
related to institutional design and effectiveness i.e. Objectives 8, 9, 10, 11, 12 and 13.
We adopt the goals focused approach to measuring the remaining objectives. This approach
requires understanding of the hierarchy from goal to objectives to objective measurement and then
technical or perception measurement. Table 4-4 is a breakdown of the vision, objectives, objective
measurement and then perceptional measurement which we would propose for investigation in
future C2C-MSE projects.
39
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Table 4-4 Suggested measures of HW Partnerships vision and objectives for further
consideration in MSE development
Vision Objectives Objective measures Perceptional measures*
To be managed: Maintain and
enhance effective
institutional
arrangements at
all levels.
Expenditure on meetings
% of HWP staff time on
linkage meetings
List of outcomes from current
arrangements
Time and expenditure of
partners, stakeholders and
community on HWP projects
compared to time on HWP
management
Assess time & effectiveness of
community groups with HWP staff
Assess time & effectiveness of
stakeholders with HWP
Assess time & effectiveness of
partners with HWP staff
Invest in
continually
improving the
knowledge,
information base
and tools to
facilitate efficient
and effective
management.
Expenditure on information
tools
List of tools, new knowledge
and information base from
expenditure
List of publications & patents
(scientific, reports &
communication)
Compare publications &
patents over time
Assess perceived usefulness of
knowledge, tools and databases by
CEOs committee & Policy Council/
Board
Assess perceived ease of use &
application of knowledge, tools and
databases by CEOs committee &
Policy Council/ Board
Assess partners perceptions of
usefulness of tools, knowledge and
databases and gaps
Assess partners perceptions of ease
of use and application of tools,
knowledge and databases and gaps
In collaboration
with community,
government and
industry:
Conduct activities
and programs in
an efficient,
professional,
transparent and
accountable
fashion.
External audit of finances to
determine best use of
financial resources
External audit of management
structure and processes
ensure no unnecessary
duplication of effort
Assess community’s, stakeholders
and partners perception of HWP
transparency
Assess how accountable community,
stakeholders and partners are to
HWP outcomes and vice-versa
Assess if current HWP process are
efficient for community, stakeholder
and partners
Encourage and
support active
collaboration
between
community,
government and
industry to achieve
the Healthy
Waterways Vision.
List meetings and events
organised by HWP for
communities, partners and
stakeholders
External audit to critical
assess collaboration
arrangements and
enactments
Account for the amount of
external funding from
community, government and
industry
Assess from community, stakeholder
and partner perspective how
supportive HWP for:
o Achieving its vision
o Supporting them to achieve
vision
o Collaboration to achieve vision
o Asses from policy perspective
how influential HWP was in
impacting policy
Build community
understanding and
ownership of
problems and
support in
achieving effective
solutions.
Number of downloads of
education information from
website
Account for number of
educational presentations/
engagements HWP staff and
partners employed
Account for number of
community initiated ideas (i.e.
letters, phone calls and
emails regarding problems
and solutions)
Assess how perceptions of inclusive
community, stakeholders and
partners are in HWP management
planning and operations
Assess how community groups
identified and overcame problems
Survey broad community of HWP
vision elements and their
understanding of environmental
health, over time.
Engage the
community in
identifying issues
and developing
and implementing
appropriate
solutions.
Number of community
meetings/ presentations held
% HWP secretariat and
partners in community based
activities
Survey community on adoption of
water saving devices and activities
to improve water quality
Assess perceptions of community on
issues of water quality
40
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
*Assess – usually refers to a qualitative assessment such as survey style instrument using a one to
five scale question structure (Likhert scale).
As noted in Table 4-4, most measures are qualitative and will not easily fit in to the structured
modelling of MSE, further consideration of ‘soft systems’ methods need to be incorporated into MSE
for inclusion of full HW Partnership vision.
4.6. The management actions model
For the pilot study, two main families of management actions were implemented: changes in land-
use areas per sub-catchment and remedial constituent-transport management per sub-catchment.
The first family of actions allows us to simulate land-use changes over time such as population
growth taking up more space or changes from plantation forests to agriculture.
Figure 4-9 A screen shot of the user interface that supports the entry of n management
actions.
Figure 4-9 shows a screen shot of the interface to support a land-use-based management lever. The
underlying management action model applies the land-use changes to the response model between
the from-date and the to-date as specified (left, top) in a linear fashion. For instance, if a land-use
scenario specifies a change to 100% greenspace between 01/01/1995 and 01/01/2000, the
catchment land-uses on 01/01/1995 are gradually (linear) changed between those two dates to
result in the requested land-use configuration in the year 2000.
The second family of actions allows specification of remediation activities, such as a change in the
constituent runoff through water-sensitive urban design measures to human habitation land-uses
within a sub-catchment or the restoration of riparian vegetation in agricultural parts of a sub-
catchment.
41
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 4-10 The remediation activities entry screen.
Figure 4-10 shows a screen shot of the interface for specifying remediation actions. These remedial
actions are defined as input-output transfer functions (graph on the left hand side) for each of the
constituents. This approach was inspired by the generic node approach of the EWater CRC
stormwater management package MUSIC. The left-hand side allows the specification of the transfer
function to be applied to the flow and each of the constituents. It also allows the specification of
which land-uses to apply these remedial actions to. The right-hand map allows the user to specify
which sub-catchments will be affected by the remedial management actions.
The land-use and remedial management levers are not only needed to (manually) enter the various
management scenarios, but they also form the basis for the management decision model. The
existence of these management levers allows the management decision model to change
management action in an adaptive fashion, in cases where management decision rules are
available.
4.7. The Management Decision and Learning models
In the Catchment-to-Coast MSE context, a management decision is the choice for a set of
management actions. The goal that drives a decision is the discrepancy between the current status
of the system under management and its management objectives. In general, we’d use information
(knowledge, assumptions) about the system that we manage AND our expectation of the
effectiveness of the management levers we have to our disposal to select a set of management
actions out of a much broader package of possible management actions. We refer to a method that
we are using to combine goal, information, assumptions into a decision as a decision strategy.
Examples of a decision strategy are a completely random choice of management actions (effectively,
we do not learn and adapt), or some ‘rules of thumb’ to manage resources (‘if the result of the
assessment is this, than we do that’) or even a sophisticated technique using the optimal control
approaches (Lewis and Vassilis 1995).
Learning is the process of adapting our information, assumptions and even strategies between the
iterations of an adaptive management scheme. Such learning is often based on the experiences
collected from the previous decisions and the actual assessments of their effects.
In the pilot project, the Management Decision and Learning models are not yet implemented in the
pilot system. The main reason is the relatively complexity of these approaches and the unavailability
42
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
of explicitly stated management rules. A separate project is now underway to research the details
required to implement these aims.
4.8. Summary
In this chapter we described the models implemented in the HW-MSE pilot application. The
catchment model plays a major role in the pilot because it generates the main factors (flow and
constituents) that form the base of the existing monitoring and assessment programs for water-
related issues in SE-Queensland. To a certain extent, this reflects the main focus of research,
monitoring and assessment over the last 10 years. It is also widely recognised that that may not be
the appropriate base for resource management scenario evaluation in the future. To open up the
HW-MSE to include other areas (e.g. economics, social, institutional arrangements) some
preliminary work has been included in the model in the form of the aquaculture farm economics and
its impacts on the Queensland State level. The derived employment estimates, direct and indirect,
are examples of social performance measures that can be assessed in the pilot MSE system.
Already a program of monitoring key economics in SE-Queensland is underway. Some preliminary
thoughts on how to include institutional arrangements in the next iteration of the C2C-MSE have also
been discussed in this chapter.
43
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
44
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
5. THE LOGAN-ALBERT PILOT MSE SOFTWARE
This chapter will discuss in broad terms the functionality of the software package that was developed
in the pilot project to support the C2C-MSE activities. The last section of this chapter will present
some example results of the Logan/Albert pilot MSE.
5.1. HW-MSE application preliminaries
In general, performing an MSE is a relatively complicated, often time consuming and drawn out
process that sometimes takes months or even years. A major part of the work is interacting with
stakeholders on management issues and objectives, system understanding and management
constraints and so on. On a more technical level, we need to collect key data, set up data storages,
decide on the contents of each of the management activities (see Figure 2-1) and implement these
activities (either in automated or manual form). Once this is in place, often many sessions with
stakeholders are needed before the MSE is accepted as a management tool. The current pilot
software system supports only part of these activities:
1. Projects: delineating models, data, functions, scenarios and MSEs in projects.
2. MSE specification: parameterisation of the available (stochastic) models through a range of
general and custom-made interfaces, combining models for each of the management
activities (Figure 2-1) into scenarios and grouping scenarios into a MSE.
3. Evaluating one or more scenarios.
4. Monitor progress and results.
5. Examining MSE results on two levels: management-support tables summarise results, and a
relatively flexible approach to combine time series data resulting from the MSE scenario
evaluations.
6. A small set of software maintenance tasks such as adding software services and inserting
new model parameters.
7. Structuring models for use in the MSE system (plug and play standards).
Apart from the quite visible tasks that can be controlled through the user interface, there are a range
of tasks being performed in the background such as the implementation and execution of
communication standards and general services such as managing data flows between models and
data storages, providing mathematical, statistical and GIS functionality, linking between the MSE
software and other packages such as R and Matlab etc. See Appendix A for more details on the
software development.
5.2. HW-MSE application workflow
In the section we will briefly discuss the four main activities that are supported by the pilot software.
They are Specification, Evaluation, Monitoring, Results and Post-processing. Two others, Projects
and Maintenance are mainly technical support tasks and will not be discussed here. This section
aims to give a feel for the functionality of the pilot software and is not meant to be a user manual.
5.2.1. Specification
The Specification tab (top of Figure 5-1) is central to defining the models that constitute a scenario
and the scenarios that constitute a MSE. It allows the choice of models that populate each of the
management elements (six coloured boxes in Figure 5-1).
45
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 5-1 The specification screen allows creating different models and construct
scenarios.
The specification screen has been designed to reflect our thoughts regarding the management
stages, as can be seen in Figure 5-1. Each of the six coloured boxes contains one or more models
that represent that management stage. For the discussion of the software functionality, a ‘model’
consists of a named set of software (code), data and parameter (with their values). If we change
(and save) parameter values of a model, we say that we have created a new model. For instance, if
we were to change the land-uses of some catchment model, the software will allow us to save those
changes as a new model. The list in the response model box in Figure 5-1 shows four models that
we can use to compose a MSE scenario. Selecting a model and clicking on the big coloured box
gives access to the model’s parameters. Once changed, they can be saved under another name,
thus creating a new model. The software supports the stochastic running of models, so a full set of
stochastic descriptors regarding their statistical distribution is available for each parameter.
Models may have many thousands of parameters, often with hierarchical relations, and a concerted
effort has been made to implement a range of interfaces to those parameters. For instance, Figure
5-2 shows a screen to access the nearly 5,000 parameters of the SIMHYD catchment model (Chiew
et al., 2002). The map on the left of Figure 5-2 allows the selection of a sub-catchment and brings
up a screen with all land-uses (tabs on the right-hand) and parameters (Excel-like data grid). Each
parameter has a range of descriptors, of which Figure 5-2 shows only a subset, such as its value,
unit and description needed for stochastic runs. The stochastic descriptors for each of the
parameters allow us to specify from what distribution the parameter should be drawn during
stochastic runs to the assess uncertainty in the results.
46
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 5-2 Models may have thousands of parameters. Special-purpose screens are available
to support easy access to those parameters and make editing less error-prone.
By designing appropriate interfaces and making the process of exploring various options easier and
less error-prone, we expect that the user will explore a wider range of options. In complex
environments such as multiple use resource management, exploring a wide range of options is likely
to help find better solutions.
Once we have designed a collection of models for each of the MSE elements, the next step is to
compose a scenario using a subset of the models that we created. Figure 5-3 shows an interface
that allows us to create a MSE, and compose one or more scenarios and add them to the MSE.
Figure 5-3 A MSE is a collection of scenarios. Each scenario consists of some administrative
information and a range of models, representing the major management elements.
47
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Composing a management scenario is a simple task once the underlying models have been
generated. The drop-down lists in each of the now familiar management elements (six coloured
boxes in the middle of Figure 5-3) contains the available models. Once a selection of the models
that comprise the scenario have been made, the new scenario is given a name and by saving it, it is
added to the MSE specified in the text box at the top of Figure 5-3. The next task is to evaluate the
scenarios.
5.2.2. Scenario Evaluation
Once the scenarios are composed from the models, the next activity is to evaluate them. Figure 5-4
show the interface that lets us choose which of the available scenarios to evaluate. There may be
many scenarios that need to be evaluated and some visual feedback is needed during the evaluation
(running of the models in the scenario) to keep track of what is happening and where we are in the
overall evaluation.
Figure 5-4 After composing the management scenarios, the next activity is to evaluate them.
Feedback about which scenario is currently being evaluated is provided in the green list box in
Figure 5-4. This list box also gives us some information on the time that is needed to evaluate a
scenario, allowing a rough estimation of the total length of all evaluations. The evaluation may be
stochastic (multiple runs), and the list-box also give some information on what replicate is currently
being evaluated.
The blue list box in Figure 5-4 shows information regarding the computing time spent in each model.
For instance, in the case shown in Figure 5-4, the Response model consumes around 80% of the
computing resources. This feedback is particularly important information for the modellers and
programmers to target their efforts if the evaluations are taking too much time.
During the scenario evaluation, there is another level of visual feedback of interest to the user,
namely the outputs of the models. The Monitoring tab allows us to inspect ‘on-the-run’ results.
5.2.3. Monitoring the evaluation
The Evaluation Monitor tab in Figure 5-5 shows outputs of the models that are being evaluated. The
three plotting radio buttons on the ‘Evaluate scenarios’ tab (Figure 5-4) allow us to specify some of
the behaviour of the monitoring plots, whether to update the plot for every time step or to wait until
one full evaluation has been performed. As the yellow tree-view on the left side of Figure 5-5 shows,
we will be able to select the main dynamics of each of the models in the active scenario for
monitoring.
48
Healthy Waterway Management Strategy Evaluation – Science Support for catchment-to-coast water quality management
Figure 5-5 The Monitor tab allows inspecting the results of various models.
For instance, the evaluation monitor tab in Figure 5-5 shows that information is available from the
Response, Observer and Assessor models (three tabs on the bottom of the graph). The three traces
on the graph show the concentrations of total nitrogen and total phosphate and water turbidity of the
Observer (note the uncertainty boxes in the time series). The Monitor allows us to follow the scenario
over time and stochastic replication by