Savanna riparian health

Leader: Michael Douglas, Charles Darwin University

Full title: Savanna riparian health: definition, indicators and response to management
Project 1.2.1

Summary | Activity 1: Framework for riparian research in the Australian savannas | Activity 2: Assessing and managing the health of riparian vegetationActivity 3: Riparian health and stream vulnerability | Activity 4: Hydrological processes and riparian health | Activity 5: Fluvial indicators of riparian health | Activity 6: Remotely sensed indicators of riparian health | Activity 7: Weed management and riparian health | Outcomes | Outputs | Project team |


The Savanna Riparian Health (SRH) project is a collaborative venture between partners in the Northern Territory, Queensland and Western Australia. Major contributors are the Charles Darwin University, James Cook University and CSIRO with significant support from Commonwealth, State and Territory agencies.

Riparian zones are vital elements of the savanna landscape. Their contribution to biodiversity, cultural values and the economy is disproportionate to the small area they occupy. They are important to maintain savanna and in-stream biodiversity, stream channel morphology and water quality. However, savanna riparian zones are highly vulnerable to the effects of disturbances such as weed invasion, feral animals, fire and overgrazing. Threats to riparian health are compounded by the fact that riparian zones are the focus for much activity related to the development of northern Australia (such as grazing, agriculture and tourism) and the concentration of use in these habitats is likely to increase in the future. There is therefore an increasing need for practical management options in riparian zones that will allow savanna landholders to meet their management objectives while still maintaining riparian health.

Developing practical management options for riparian zones requires: (1) a definition of riparian health and, (2) practical methods to assess it; (3) an understanding of the effects of potentially threatening processes on riparian health, and; (4) an evaluation of techniques to manage those threats. This project addresses all four of these requirements.

Although there has been a good deal of recent work on some aspects of riparian ecology and management in tropical savannas, there is a need to draw this information together and to develop a conceptual framework that reviews our understanding of savanna riparian zones, attempts to define the concept of riparian health and identifies knowledge gaps and opportunities for future research. A large focus of the early activity of this project will be the development of such a framework. This will be invaluable for guiding future activities of this project.

Developing indicators of savanna health is a primary objective of the TS-CRC. Progress has been made for the savanna woodlands and open forest, but little research has focused on savanna riparian zones. Indicators of riparian health may be based on the vegetation itself on hydrologic or geomorphological features, water quality characteristics or other in-stream ecosystem processes or combinations of these features. This project will test a suite of possible indicators of riparian health in an effort to develop indicators that can be applied by savanna landholders. Indicators developed in one region will be tested in other regions to test their applicability at the broadest scale.

Initially, conceptual framework for understanding the nature of riparian systems will be developed. Subsequently, study sites in key catchments will focus on the development of measures of riparian health. These will principally relate to both stream and bank processes and condition.

The project is made up of seven separate but complementary activities.

Activity 1

Framework for riparian research in the Australian savannas (Burrows, Butler and Lukacs, ACTFR-JCU)

The riparian systems and their fundamental process in Australian savanna are poorly understood. Conceptual models developed in temperate regions of Australia and overseas are not appropriate and a framework is needed to define riparian health and to underpin proposed research.

Initially, an information review will be undertaken. Subsequently, a discussion paper will be produced to workshop concepts between riparian researchers. The project aims to produce a publication that defines a framework for riparian health; characterises riparian systems in the savannas across northern Australia; and supports research and management.

Activity 2

Assessing and managing the health of riparian vegetation (Michael Douglas, Samantha Setterfield, Sally Kent, (CDU), Rod Kennett, Simon O’Connor (PAN); Neil Pettit (JCU).

There are several widespread activities throughout the savannas that threaten the health of riparian vegetation (e.g. fire, weed invasion feral animals, grazing). Many of the important riparian fruiting trees are long lived and so the impact of disturbance on recruitment processes may go unnoticed for a long time. There is a need to: (a) develop methods to assess the health of riparian forests, (b) identify the factors threaten plant recruitment and (c) develop and assess practical management techniques to control or mitigate these threats.

This project will focus on two areas with contrasting land tenure and management objectives: Kakadu National Park and the Victoria River catchment in the Northern Territory. Assessment methods developed in the project will also be tested at sites in the Burdekin River catchment in Queensland.

An index of health, based on stand structure (ratio of adult to seedlings/saplings) will be developed and tested as a measure of riparian vegetation health. Several key plant species of importance (to be determined by their ecological and cultural significance) will be studied in detail to determine the role of major threatening processes in limiting their recruitment.

This will be determined by sowing experiments under a range of disturbance regimes and monitoring seedling survival. Management practices will be assessed on sites with known and contrasting fire histories, feral animal and weed control strategies and stock access fencing. In addition, experiments will be conducted using animal exclosures, controlled burning and weed control to assess management practices.

Activity 3

Riparian health and stream vulnerability (Neil Pettit, Damien Burrows, Barry Butler (ACTFR-JCU); Murray Whitehead (QEPA); Martin McLaughlin (QPWS), Peter Gilbey (QDNRM); Michael Douglas (CDU)

Previous ACTFR research has shown significant variability in the water quality and riparian health of savanna streams, both intra & inter seasonally. Key drivers appear to be biophysical, climatic and land-use related. Risk of degradation is best managed via 1) the identification of the inherent vulnerability of different stream (reach) types and 2) integrated responses (eg riparian use defined by spatial/temporal conditions).

1) Reach types (in principally Burdekin River catchment but also maybe extended to VRD catchment) will be classified utilising a GIS platform built on remotely sensed and field data. Hydrological data (eg flow, detention time), geomorphic data (eg landform & elements) and biological data (eg riparian vegetation, substrate), together with stream process data (eg benthic respiration, food webs), will be gathered & utilised to derive the classification.

2) Criteria for a vulnerability index will relate to biophysical attributes (eg diel oxygen cycling), land use attributes (eg grazing regime) and (possibly) socio-economic attributes (eg feasibility of management initiatives) of the reaches and the catchment.

3) Aside from scientific publications etc, tools will be developed for self-assessment by landholders of the inherent vulnerability of their own streams.

Activity 4

Hydrological processes and riparian health (David Post CSIRO, Mirjam Alewinjse, Robin Thwaites JCU)

It is our hypothesis that reductions in infiltration due to cattle grazing has led to a decrease in baseflow in streams in the Burdekin catchment. We expect this to be particularly true for small streams, with the effect declining as scale increases. These decreases in baseflow would have impacts on riparian ecology.

In order to test this hypothesis we are examining the infiltration characteristics of 12 cattle proof exclosures that have had very little grazing on them for approximately 15 years, and comparing the results to areas outside the exclosures that have had ‘normal’ amounts of cattle grazing on them. These results will then be input into an appropriate physically-based small-catchment model in order to observe the effect on baseflow at various scales. The TOPOG-YIELD model is one such model being considered.

Twelve cattle proof / kangaroo proof exclosures were constructed by the Department of Primary Industries (DPI) around 15 years ago in order to observe changes in runoff. Since that time, most of the exclosures have remained in operation, although all have had cattle in them at some time or another.

This project seeks to examine the infiltration characteristics of the soil inside the exclosures and compare it to the area outside them. A hood permeameter and a ring infiltromemeter will be used to determine both surface and near-surface infiltration rates both inside and outside the exclosures. Two replications of each type of measurement will be carried out – 48 infiltration measurements in total.

Activity 5

Fluvial indicators of riparian health (Karl-Heinz Wyrwoll, UWA)

This activity will link biological and water quality indicators of riparian health with geomorphologic indicators of channel form and bank stability. These indicators may provide integrated responses over longer temporal scales than many biological indicators.

Activity 6

Remotely sensed indicators of riparian health (Stuart Phinn, UQ, Bob Karfs NTDLPE)

This project will use high-resolution remotely sensed information to test indicators of land condition and vegetation health. These indicators may provide useful informationover larger spatial scales than the other indicators tested.

Activity 7

Weed management and riparian health (Tony Grice, Roger Lawes CSIRO)

Weeds are unevenly distributed across landscapes. Especially in rangeland environments, many weed species are most abundant or even restricted to riparian zones. The abundance and diversity of weed species in riparian zones are major determinants of riparian health. This activity of the project will document patterns of weed abundance within the Burdekin River catchment and relate them to riparian health and management.

This activity will link with work being done within the CRC for Australian Weed Management. It will add to that work by using skills in spatial analysis and ecology to describe patterns of weed distribution at a broad landscape/catchment scale within the Burdekin Catchment. These patterns can be related to other aspects of riparian structure, function and health. The data will be analysed to explore the processes underlying the patterns and will be reviewed to propose management approaches appropriate to various categories of weeds in riparian zones.


The Riparian Health projects will deliver outcomes in three key areas:

  • Improved understanding of riparian processes in the Australian savanna
  • A variety of tools for the assessment, monitoring and management of riparian health
  • A more informed and aware community of the value of riparian systems
  • Understanding – fundamental knowledge of the nature and behaviour of riparian systems will be generated. This material will not only enhance the existing scientific information base but also underpin the sustainable development of the savanna. Process information will be utilised in the development of management and research programs as well as educational and community awareness programs.
  • Tools – a variety of tools for the management of savanna will be developed. These will be tailored to suit a variety of stakeholders – including landholders. Tools will provide for the rapid identification of the health of riparian systems through the use of novel indicators and indices. Additionally, tools will be generated to prioritise management strategies for agencies and community in the rehabilitation of riparian systems.

Information – material for consumption by a variety of stakeholders will be developed. This will principally be through CRC extension activities and will result in communities across the savanna (principally NT and QLD) being exposed to regionally relevant information and having direct contact with researchers. A more aware and informed community will be developed that recognises the importance of riparian systems.


  • CRC publication that characterises riparian systems in the savannas across northern Australia and provides a framework for research and management
  • robust indicators that can be used to assess health of riparian systems in tropical savannas
  • methodology for a savanna-wide assessment of riparian health tested in representative catchments
  • practical options for managing riparian systems
  • CRC publication of guidelines for best practice management of riparian systems
  • post-graduate theses and scientific publications

Project team

B. Butler, ACTFR (JCU)
D. Burrrows, ACTFR (JCU)
N.Pettit, ACTFR (JCU)
M.Whitehead, QEPA
M. McLaughlin, QNPWS
P.Gilbey, QDNRM
M. Douglas, CDU
S. Setterfield, CDU
S. Kent, CDU
R. Kennett, PAN
S. O’Connor, PAN
K-H. Wyrwoll, UWA
T. Grice, CSIRO SE
D. Post, CSIRO L&W
A. Hawdon, CSIRO L&W
J. Kemei, CSIRO L&W
M. Alewinjse, JCU
R. Karfs, DLPE


Mr Damian Burrows
Research Officer
Australian Centre for Tropical Freshwater Research
Tel: 07 4781 4262

Fax: 07 4781 5589

James Cook University

A/Prof Michael Douglas
Charles Darwin University
Tel: 08 8946 7261

Mobile: 0408 467 000
Fax: 08 8946 6847

Charles Darwin University


A/Prof Michael Douglas
Charles Darwin University
Tel: 08 8946 7261

Mobile: 0408 467 000
Fax: 08 8946 6847

Charles Darwin University