Figure 1 Location of the Six Pilot Study Catchments in Brisbane City Council Area.
As of January 2004 au.riversinfo.org is archived under Precision Info
RIVERSINFO AUSTRALIA ARCHIVE
This archive provides non stylised versions of selected reference documents (formerly) resident on au.riversinfo.org.
Foreword & Publication Information
Executive Summary, Abbreviations
List of Figures
List of Tables
This report describes the outcomes of a research project conducted under the Urban Research and Development sub-program of the National River Health Program (NRHP).
The NRHP is an on-going national program established in 1993, managed by the Land and Water Resources Research and Development Corporation (LWRRDC) and Environment Australia. Its mission is to improve the management of Australia's rivers and floodplains for their long-term health and ecological sustainability, through the following goals:
Urban streams and estuaries (i.e. those affected by runoff and discharges from urban areas) are an important subset of Australia's waterways. Most are degraded biologically, physically and chemically and therefore require appropriate methods to be developed for health assessment and management. The Urban R&D Sub-program, managed by the Water Services Association of Australia, comprises 8 research projects which were developed to meet research priorities for urban streams and estuaries within the goals of the NRHP and to complement existing NRHP projects on non-urban rivers. Thus, research focuses on development of standardised methods for assessing the ecological health of urban streams and estuaries which can be linked with data on water and sediment quality. The urban R&D projects commenced in 1996.
The definition of health in urban waterways used is "the ability to support and maintain a balanced, integrative, adaptive community of organisms having a species composition, diversity and functional organisation as comparable as practicable to that of natural habitats of the region".
The eight projects of the Urban Sub-Program are:
| Decision support system for management of urban streams | Dr John Anderson Southern Cross University, Lismore |
| RIVPACS (River InVertebrate Prediction and Classification System) for urban streams | Dr Peter Breen CRC for Freshwater Ecology, Monash University, Melbourne |
| DIPACS (Diatom Prediction and Classification System) for urban streams | Dr Jacob John
Curtin University, Perth |
| Sediment chemistry- macroinvertebrate fauna relationships in urban streams | Dr Nick O'Connor Water EcoScience, Melbourne |
| Classification of estuaries | Dr Peter Saenger Southern Cross University,Lismore |
| Literature review of ecological health assessment in estuaries | Mr David Deeley
Murdoch University, Perth |
| Estuarine health assessment using benthic macrofauna | Dr Gary Poore
Museum of Victoria, Melbourne |
| Estuarine eutrophication models | Dr John Parslow CSIRO Marine Laboratories, Hobart |
Basic Decision Support System for Management of Urban Streams Part C Pilot Studies in Brisbane
Dr. John R. Anderson
Centre for Coastal Management Southern Cross University
LWRRDC Occasional Paper 10/99 (Urban Subprogram Report No. 3)
December 1999
ISSN: 1320-0992
ISBN: 0-642-267760-X
Published by:
Land and Water Resources Research and Development Corporation
GPO Box 2182 Canberra ACT 2601
Telephone: (02) 6257 3379
Facsimile: (02) 6257 3420
Email: public@Iwwrrdc.gov.au
WebSite: www.lwrrdc.gov.au
© LWRRDC
Published Electronically on au.riversinfo.org by the Environmental Information Association (Incorporated) with the permission of LWRRDC and Environment Australia. Environment Australia assisted by providing copies of the manuscript for electronic publication. The Natural Heritage Trust provided project funds which were used to assist in publishing this material. In the case of variation between this document and the hard copy original the original takes precedence. (Bryan Hall).
The information contained in this publication has been published by LWRRDC to assist public knowledge and discussion and to help improve the sustainable management of land, water and vegetation. Where technical information has been prepared by or contributed by authors external to the Corporation, readers should contact the author(s), and conduct their own enquiries, before making use of that information.
The aim of this study was to develop a software based classification system for management of urban waterways. A decision support system was required that used biologically important physical attributes to classify the streams and waterways in terms of asset value and capability for rehabilitation or enhancement, physical and environmental condition and key constraints limiting restoration. Four reports have been published for this study. The development of the basic 'Decision Support System for Management of Urban Streams' is described by Anderson (1999a), the software system by Anderson (1999b) and a demonstration package in the form of a game 'Let's Fix Urban Streams' by Anderson (1999c). This report describes the pilot studies in Brisbane.
The approach adopted was to develop a series of classifications rather than a single integrated classification system. The method is based on rapid surveys by 'non-expert' personnel who are trained for the study. The sampling is based on defining 'homogeneous' stream sections of varying length. The physical and ecological characteristics of the habitat within these sections and the bed, bank and shoreline buffer zones are similar throughout these sections. One or more sites, representative of these sections, are surveyed and the information is applied to the entire length of the section. The site consists of a reach of varying length, usually about 50m long. The local drainage area for each section is defined. The subdivision of the streams into homogeneous stream sections and their associated catchment elements is continued to cover the entire length of stream in the drainage network and the entire catchment area within the watershed. The site (point), stream section (arc) and local catchment (polygons) form the basic GIS element. A large integrated database system is developed for the sites and sections. A software package is used to produce a series of condition ratings and various other summary parameters that characterise the channel and shoreline buffer areas in terms of ecological and physical type, dimensions, condition and extent of modifications. These derived parameters are then stored in the database system, which provides a basic decision support system for managing urban streams. A comprehensive set of reports, data summaries and mapping procedures are available to assist the user in providing the type of information required. The key components of the system are:
This report concerns the pilot studies for the project that were conducted in Brisbane in 1996-1998 to develop and refine the survey methods, data analysis procedures and software system for the decision support system. The pilot studies were conducted in association with the Brisbane City Council, who provided extra resources to enable the pilot study to be expanded to a complete evaluation exercise of six subcatchments in Brisbane:
| Catchment | Catchment Area (sq. km) | Length of Major Channel Surveyed (km) | Sites | Sections |
| Norman Creek | 28.6 | 32.1 | 79 | 86 |
| Moggill Creek | 57.6 | 42.7 | 85 | 103 |
| Bald Hills Creek | 8.5 | 16.0 | 20 | 33 |
| Nundah/ Downfall Creek | 35.0 | 42.7 | 75 | 86 |
| Enoggera/ Breakfast Creek | 89.2 | 80.6 | 125 | 181 |
| Kedron Brook | 110.0 | 42.9 | 90 | 102 |
| Total | 328.9 | 257.0 | 474 | 591 |
The pilot studies were used to develop and refine the datasheets, database design, software system, report and data summary procedures and the basic decision support system.
The system was developed to be used by 'non-experts' who are trained for the surveys and to use the software system. Community groups were involved in all phases of the project including the planning, reconnaissance and surveys. Brisbane City Council and one of the community groups also trialed the software package. Their feedback was used to develop and refine the package.
The results of the pilot studies are discussed using the various reporting facilities available in the system to illustrate how the system can be used to audit the condition of the streams and to classify various sections in terms of habitat types, dimensions and other key features useful for management. Various 'Skeleton' maps have also been produced to shown the distribution of some of the key attributes throughout the pilot catchments.
The pilot studies were evaluated and used to provide estimates of the time and personnel required for similar studies. These estimates are provided for a catchment requiring 200 sites, but additional information is provided to revise these estimates for smaller or larger catchments.
These estimates show that the methods are rapid, cheap and cost-effective. They provide a comprehensive database system and classification procedures. The basic decision support system is versatile and powerful allowing a wide range of management issues to be easily addressed. The condition of the stream sections within a catchment and their various features can be easily rated and ranked in various categories and combinations that are defined by the user. The database system means that all the information can be examined and used for management decisions. The surveys are designed to be repeated to show trends and rates of change in condition or other channel features. Follow-up surveys can also be used to periodically monitor the outcome of various management initiatives.
Abbreviations
The overall aim of this project was to develop software based classification system for management of urban waterways. A decision support system (DSS) was required that used biologically important physical attributes to classify the streams and waterways in terms of asset value and capability for rehabilitation or enhancement, physical and environmental condition and key constraints limiting restoration.
The development of the classification system has been described in Anderson (1999a) and the software and manual by Anderson (1999b). These methods were developed from similar approaches used to assess the ecological and physical condition of rural waterways. The 'State of the Rivers Methodology' and the 'Riverine Habitat Audit Procedures' are described by Anderson (1993 a, b, c). These methods have been applied to 20 major catchments in New South Wales and Victoria. Various reports have been published on these methods (Carter 1997; Moller 1996; Phillips and Moller, 1995; Telfer 1995).
Pilots studies of the decision support system were conducted in Brisbane in association with the Brisbane City Council (BCC) to:
This report describes and discusses the pilot studies that were conducted from late 1996 to 1998.
It was originally intended that only a single catchment in Brisbane (or part of a catchment), would be surveyed as a pilot study. However Brisbane City Council (BCC) was very enthusiastic about the project particularly in using it to prepare catchment management plans. BCC provided extra funds and staff time to expand the pilot studies to a full trial of the survey techniques and software system in a major part of Brisbane. These extra resources and the efficiency of the rapid survey methods meant that full and complete trials could be completed for six major catchments, within a total catchment area of about 330 square kilometers. This greatly increased the scope of the trials from pilot studies to comprehensive studies at the scale of resolution required for final complete studies of the catchments. This provided a more extensive test of the methodology and also provided a more useful outcome than a simple pilot study. Brisbane City Council subsequently conducted surveys of a further six catchments in 1998 using these methods, but reviews of these additional surveys were not included in this report.
Feedback from BCC and the community groups during the pilot studies was used in an ongoing process to upgrade and modify the procedures, and software systems. All the databases, maps and reporting procedures for all catchments were upgraded to the final form before being made available to BCC and community groups at the end of the project. Additional comments on the methodologies and software system are welcome. Please contact the author. These comments will be used in developing future upgrades of the package.
The development of the survey methods and datasheet has been described by Anderson (1999a). The datasheets were extensively discussed and reviewed during the workshop held just prior to the pilot surveys, which were conducted in December 1996. Further refinement was made during the pilot studies using feedback from the survey teams and BCC. Revised datasheets were prepared in response to comments and suggestions and the final versions of the data sheets and databases reflect these changes.
Preliminary discussion with Brisbane City Council and evaluation of the resources available and required for the study suggested that about 500 sites could be surveyed for the pilot study. The catchments to be included in the surveys were selected by Brisbane City Council based on their need for data to prepare catchment management reports and the interest and support of local community groups. The involvement of local community groups was a high priority to generate a sense of community ownership for the survey data and to test their participation in the surveys and use of the classification and software system. Catchment size was also important in determining how many catchments could be surveyed, and it was decided that a mixture of small and large catchments would be included to test the methods. The six catchments finally chosen for the pilot studies are listed below in order of decreasing catchment area:
Kedron Brook is the second largest catchment in Brisbane (110 sq. km), and Enoggera/Breakfast Creek (89.2 sq. km) is also relatively large. Bald Hills (8.5 sq. km) and Norman Creek (28.6 sq. km) are relatively small catchments. The choice of a range of catchment sizes was seen as being important for the pilot study. Enoggera, Moggill and Norman Creeks drain into Brisbane River, Bald Hills Creek into Pine River and Kedron Brook and Nundah/Downfall Creek drain directly into Moreton Bay. Figure 1 shows the location of the pilot study catchments. Maps for each of the pilot catchments are shown in Figs 2-7.
|
Figure 1 Location of the Six Pilot Study Catchments in Brisbane City Council Area. |
| (original size image) |
The support of local community groups was essential for choosing of the catchments to be included in the pilot study. Community groups were an integral part of the reconnaissance surveys and the pre-survey planning for the project. These groups provided invaluable local knowledge to help identify the boundaries for the 'homogeneous' stream sections and to select representative sampling sites for each section.
The availability of community volunteers for the surveys was also important. Given the need to train the survey teams, the following criteria were established for involvement of community volunteers in the surveys:
Pre-survey planning workshops, were organized for each of the catchments. This involved BCC staff and community groups. The purpose of these workshops was to review all of the existing information to define the 'homogeneous' stream sections in a preliminary way as a mapping exercise. Waterway planning maps and maps of land use and vegetation cover and BCC's 'BIMAP' GIS system were used for this exercise. The local knowledge of the community groups was invaluable for this process. The resources available ultimately limited the number of sites that could be surveyed in total and for each catchment. The preliminary allocation was an average of one section for each kilometer of stream length. This allocation was two to three times the site allocation previously used for rural catchments (average of one site for every 2-3 km. More sites and smaller sections were needed in urban areas because of the greater complexity and extent of channel modifications in urban areas. The length of the sections varied in different parts of the catchment. More sites and sections were generally required in the headwaters because of the higher branching and stream density for the lower order streams. The lowland river sections towards the mouth of the stream generally required fewer sites and the sections, because there were fewer tributaries and long lengths of stream have similar characteristics in these areas. Sections were up to 5km long in these areas. A preliminary set of sections was defined as a mapping exercise using the workshops. The section boundaries were transferred to 1:10,000 working maps for the catchments.
Reconnaissance surveys were then conducted for each of the catchments to refine the section boundaries and to select representative survey sites. This required one person to visually inspect all the river sections, to check their homogeneity, to refine the section boundaries and to select the sites and reaches to be surveyed. Community groups were used to provide local input for this process. Several people were involved in different parts of the catchment. The 'Site Description' data sheet was completed with a description of the location of the site and survey reach. Most of the photographs were also taken during the reconnaissance surveys to reduce the workload for the full survey team and to reduce the number of cameras needed for the surveys. GPS and map coordinates were also completed for each site using local street directories and topographic maps. These ensured that the full survey teams were able to locate each survey site quickly and reliably. These reconnaissance surveys in the six pilot catchments required 10 days, with an average of about 20 sites a day being completed by a single person.
Once the reconnaissance surveys were completed the major survey exercise was planned. Five teams of two people were required for two weeks and for the two-day training exercise, which preceded it. Eight to ten sites were allocated for each of the teams providing a total potential of 450-500 sites, including the second training day when about 4 sites were surveyed. Two of the sites were designated as lower priority 'bonus' sites. The survey teams were encouraged to keep on schedule by completing all of the sites allocated for each day. Mobile phones were used to coordinate the surveys. The project leader periodically visited each team to deal with problems as they arose and to check the data sheets were being completed correctly.
Five teams of two persons were established for the surveys. The personnel consisted of:
There was some interchange of personnel due to other commitments, but each member was involved for at least 5 days.
Each team acted independently to complete the sites allocated to them within the two-week survey period. The community group volunteers were covered by insurance and were provided with a daily allowance to cover expenses such as lunches, sunburn cream, insect repellent etc. The University students were employed temporarily for the surveys.
Each team of two was provided with their own vehicle and equipment. Four-wheel drive vehicles were hired for the surveys. The survey equipment needed is described by Anderson (1999a). Each team was provided with eight sets of data sheets printed on waterproof paper (one days supply) to enable the surveys to continue during showers and light rain. The surveys had to be postponed during moderate to heavy rain.
Surveys of water quality and common aquatic and riparian weeds were organized as supplementary surveys. A weed identification seminar was conducted by BCC staff during the training workshop using plant specimens and photographs. Each team was provided with a set of photographs and keys to identify each of the plants selected for inclusion in the weed list. A supplementary datasheet was included to record this information. The data base system allows for additional plant species to be included in the checklist on the 'Vegetation' data sheet.
Water quality survey kits ('Water watch') were also provided for each team to enable simple water quality measurements to be taken during the surveys. These tests included temperature, electrical conductivity and turbidity (using Perspex tubes). The existing data sheets allowed this extra information to be recorded. The survey methods already included a visual assessment of water quality problems at the site.
A two-day training workshop was held as described by Anderson (1999a). A half-day seminar was used to introduce the survey methods and data sheets. The emphasis was on 'learn-by-doing' with trial sites being surveyed under supervision in the afternoon of the first day. Actual surveys began on the second day with each team being allocated four sites. An experienced person supervised and assisted each of the teams for at least two of these sites. All the data sheets completed during the second day were checked on site. A de-briefing session was held at the end of the second day to address any problems with the data sheets, methods and equipment.
The five teams of two persons conducted the surveys during the first two weeks of December 1996. Each team acted independently. Two teams continued for a further week to finish off the surveys and to survey many of the 'bonus' sites originally allocated.
The expansion of the pilot surveys to six catchments and 509 sites created a major task in entering and verifying the data into the databases. This caused major delays in the completion of the project. The data entry is best completed by someone who has experience with the survey methods, preferably someone who was actually involved in the surveys. This experienced person can correct errors and inconsistencies as they arise using the photographs and so ensure that the data entered is complete and accurate. It is important to ensure that such an experienced person is available for these tasks. The data for Moggill, Enoggera and Bald Hills Creeks were completed first to meet BCC's needs for preparing catchment management plans. The data entry and analyses for the other three catchments were completed in 1998. The delays arose because experienced staff were not available for these tasks and the data entry forms and software programs were being developed simultaneously with the data entry.
The software package includes a set of checking and verification programs that are essential for checking the data prior to analysis.
The software package was developed and refined by consultation with BCC staff and community group personnel who were trained in its operation. These personnel provided feedback on the package. Full reports and draft maps were also provided for comment to finalize the outputs and data summaries produced by the package. Various data summaries and GIS outputs were also provided to BCC for inclusion in preparing catchment management plans and other reports. The 'skeleton map' package (Anderson 1998a) was developed in 1998 in response to the delays and high cost of GIS for the catchments. This package was needed for exploratory analyses, for providing 'instantaneous' decision support for the user and for preparing draft reports and maps. This was not seen as a replacement for GIS for production of the final output but as a simple, cheap and quick method of generating output while the GIS maps were being finalized. The 'Report Card' output (Anderson 1999a) was also produced in response to feedback from BCC staff. They needed a summary format for the 'planning units' used in Brisbane which included condition assessments as well as a variety of classification information about the channel type, extent of modifications, channel dimensions and channel habitat features.
The survey statistics for the six catchments are summarized in Table 1. Figures 2-7 show the sites and section boundaries for the six catchments. In most cases all of the more or less permanent and well-defined channels in the catchments ('main' channel) were included in the sections. The local subcatchments area associated with each of the stream sections were defined. Through this process the entire area of the pilot catchments was included. Sections of stream that passed underground through pipes were not included in the survey or were designated as 'non-survey' sections.
Only a relatively small number of sites were allocated to the natural bush land areas upstream of the dam in the Enoggera/ Breakfast Creek. Access to these natural water supply catchment areas was very difficult. The few sites surveyed probably provided a good indication of the condition and type of habitats present, as these natural areas are in very good condition. However, the sites surveyed may not adequately describe the physical dimensions and features of the channels in the area not surveyed in terms of channel width, pool depths, channel habitat types and sediments. Elsewhere the surveys provided adequate coverage of all of the waterways within the pilot catchments.
The five teams surveyed a total of 509 sites over a three-week period (Table 1). Each team completed an average of about 7-9 sites on each day, with a maximum of 15 sites. Only three days were needed to survey Bald Hills Creek catchment. The 125 sites in Enoggera/ Breakfast Creek required two teams for a total of 10 days. Additional surveys over 3-5 days were necessary to complete surveys for Norman Creek and Enoggera/Breakfast Creek.
The average allocation in terms of catchment area was 0.8-2.7 sites per square kilometer. A total of 258 km of main stream was included in the surveys from a total of 324 km identified within the area of the catchment surveyed. The length of the 'homogeneous stream sections' varied considerably within and between catchments. The lengths ranged from 20m to over 5 km, with an average of about 400m (Table 1). The longer sections generally occurred in the undisturbed upstream sections of the catchment. Longer sections also occurred far downstream, where there was little change in the channel condition over long distances. The average length of reach surveyed at the sites was about 60-90m (Table 1).
Maps of the six catchments surveyed showing the section boundaries and the site locations are provided in Figures 2-7. Usually there was only one site for each section. Generally the arbitrary site and section number codes were identical and so the site numbers shown also refer to the section numbers. The maps also show 'non-survey' sites where sets of photographs were taken but no data was collected. There were 10 of these 'photo-only' sites. There were also 70 'non-survey' areas, usually in inaccessible areas. These areas were defined as 'non-survey' sections in the databases. Data transfers were also used for some sections. The database system allows these data transfers to be made from neighbouring 'source' sections with virtually identical condition and characteristics. The sections involved in these transfers were identified during the reconnaissance and full survey exercises. This facility was only rarely used. It was mostly used where the sub-division rules for the sections required boundaries in identical areas upstream and downstream of a junction of a minor tributary junction. There were 34 sections to which data was transferred (Table 1).
| Table 1. Summary of sites, sections and section lengths for pilot surveys in six catchments in Brisbane. | |||||||
|---|---|---|---|---|---|---|---|
| Norman Creek | Moggill Creek | Bald Hills Creek | Nundah/ Downfall Creek | Enoggera /Breakfast Creek | Kedron Brook | Total | |
| Catchment Area (sq. km) | 28.6 | 57.6 | 8.5 | 35.0 | 89.2 | 110.0 | 328.9 |
| Sites (total) | 79 | 85 | 20 | 75 | 125 | 90 | 474 |
| Sites (non-survey) | 1 | 1 | 1 | 0 | 1 | 6 | 10 |
| Sites/sq. km of catchment | 2.7 | 1.5 | 2.4 | 2.0 | 2.1 | 0.8 | 1.7 |
| Sections surveyed | 86 | 103 | 33 | 86 | 181 | 102 | 591 |
| Sectionsnot surveyed | 0 | 8 | 7 | 9 | 46 | 0 | 70 |
| Sections-transferred data | 7 | 10 | 6 | 6 | 10 | 12 | 34 |
| Length of Main channel Surveyed (km) | 32.1 | 42.7 | 16.02 | 42.7 | 80.62 | 42.9 | 257.0 |
| Total length of Main Channel (km) | 32.1 | 44.3 | 19.01 | 47.9 | 111.3 | 72.9 | 327.5 |
| Average length of Surveyed Section (km) (range) | 0.37 (0.1-1.60) | 0.45 (0.1-2.0) | 0.62 (0.08-3.7) | 0.54 (0.02-5.10) | 0.59 (0.06-4.98) | 0.71 (0.10-4.0) | 0.55 (0.02-5.1) |
| Length of reach surveyed at sites (m) | 85 (15-1000) | 59 (20-400) | 62 (20-140) | 87 (30-400) | 69 (30-250) | 65 (30-250) | 74 (15-1000) |
| Days to survey | 10 | 12 | 3 | 9 | 10 | 8 | 17 |
| Teams for survey | 1 | 1 | 1 | 1 | 2 | 1 | 5 |
| Average Sites per day per team | 8 | 7 | 7 | 8 | 7 | 9 | 8 |
 |
Figure 2 Sites and Sections for Pilot Studies in Norman Creek Catchment |
| (original size image) |
 |
Figure 3 Sites and Sections for Pilot Studies in Moggill Creek Catchment |
| (original size image) |
 |
Figure 4 Sites and Sections for Pilot Studies in Bald Hills Creek Catchment |
| (original size image) |
 |
Figure 5 Sites and Sections for Pilot Studies in Nundah/ Downfall Creek Catchment |
| (original size image) |
 |
Figure 6 Sites and Sections for Pilot Studies in Kedron Brook Catchment |
| (original size image) |
 |
Figure 7 Sites and Sections for Pilot Studies in Enoggera/ Breakfast Creek Catchment |
| (original size image) |
Various reports and data summaries have been prepared for each of the pilot surveys to demonstrate the type of information that can be generated. Anderson (1999a) provides a full description of the type of data summaries and reports that can be generated from the package. A comparison of key attributes for the six pilot catchments is shown in Tables 2-7.
'Stack Diagram' summaries for the six catchments are shown in Appendix 1. These reports provide classification summaries in terms of the percentage of the total length of stream that fall within various categories in terms of condition and other attributes.
Audit summaries are shown for the six catchments in Appendix 2.
'Report Cards' for the six catchments are shown in Appendix 3. These reports for the entire catchment summarize all the key condition, channel dimension and other classification attributes in terms of the number of sites surveyed. Five report cards are shown for the first five planning areas in each catchment (two only for Bald Creek).
Four sets of 'Skeleton Maps' have been produced for each catchment showing:
The 'Skeleton' maps for each of the six catchments are provided in Appendix 4 for overall condition and riparian vegetation using the 'urban stream' set of categories (0-10%, 11-25%, 26-40%, 40-60% and >60%). Maps are also provided for Aquatic Habitat using the standard set of categories (0-20%, 21-40%, 41-60%, 61-80%, >80%), and for the channel and buffer modification index.
Six maps are provided for additional ratings and indices in Norman Creek (Appendix 5):
Various outputs have been produced for Norman Creek demonstrating the use of formulae to display combined ratings (Appendix 6). The objectives for these formulae were to identify and rank the sections for:
The use of the package for displaying maps for raw data collected during the surveys has been demonstrated for Norman Creek (Appendix 7). The attributes plotted were:
Statistical Summary Reports have been provided for the six catchments (Appendix 8).
Brisbane City Council used the data from some of the pilot studies to produce GIS maps. These maps were used for preparing catchment management plans, but they have not been included in this report.
The involvement of community groups in the reconnaissance and full surveys was highly successful. The team members performed well and their local knowledge was invaluable for planning and conducting the surveys. All participants were amazed at how their involvement in the surveys changed the way they perceived river condition and health. They also appreciated the unique opportunity to see large parts of the drainage network at such a highly detailed level. The requirement that each volunteer survey sites in neighbouring unfamiliar catchments as well as their own was also very successful. It provided a unique opportunity to see the problems and condition of neighbouring streams.
The performance of all teams was excellent. The use of University students and BCC staff as team leaders and community group volunteers as assistants was very successful as the greater technical knowledge and experience of the leaders was complemented by the local knowledge of the volunteers. It also provided a sense of local ownership for the surveys and data collected, which was considered as a major advantage in having the data accepted and used in managing the streams.
The data checking and validation procedures were invaluable for detecting error and inconsistencies in the data. The need to check and validate the data before analysis was critical. The need for an experienced person to conduct the data entry was also highlighted as many errors could be corrected during the data entry phase. It was much harder to fix and rectify data problems that showed up in the data analysis phase and in the draft reports. Careful and meticulous data entry was the key to ensuring valid outputs.
The software was developed through an on-going process by consultation with BCC and community groups who trialed various development versions throughout the project as more of the data was analyzed and became available for evaluation.
| Table 2. Comparison of Mean Condition Ratings for the Six Pilot Catchments | ||||||
|---|---|---|---|---|---|---|
| Condition Ratings | Norman Creek (78 sites) | Moggill Creek (84 sites) | Bald Hills Creek (19 sites) | Nundah / Downfall Creek (78 sites) | Enoggera/ Breakfast Creek (124 sites) | Kedron Brook (82 sites) |
| Overall | 41.4 | 44.7 | 37.1 | 36.9 | 49.9 | 41.6 |
| Environs | 38.7 | 45.0 | 33.4 | 25.6 | 44.5 | 40.9 |
| Bank | 83.2 | 56.7 | 58.1 | 73.3 | 86.5 | 89.8 |
| Bank Process | eroding | eroding | eroding | eroding | eroding | stable |
| Bed and Bar | 81.0 | 55.3 | 72.6 | 81.2 | 68.1 | 47.7 |
| Bed and bar process | stable | aggrading | eroding | aggrading | aggrading | aggrading |
| Riparian Vegetation | 12.4 | 22.3 | 9.7 | 7.1 | 24.7 | 24.2 |
| Remnant Riparian Zone width (m) | 10.6 | 15.8 | 10.4 | 5.1 | 17.6 | 17.8 |
| Aquatic Vegetation | 19.1 | 14.2 | 33.5 | 27.8 | 14.3 | 17.2 |
| Aquatic Habitat | 45.0 | 52.5 | 43.2 | 36.9 | 58.0 | 40.7 |
| Bank Cover | 49.7 | 67.4 | 59.9 | 53.3 | 78.5 | 53.5 |
| Scenic and Recreational Value | 43.2
| 60.4 | 49.4 | 42.8 | 60.1 | 45.9 |
| Conservation Value | 34.1 | 65.7 | 38.5 | 32.0 | 61.8 | 43.3 |
| Table 3 Comparison of Mean Channel Dimensions for the Six Pilot Catchments | ||||||
|---|---|---|---|---|---|---|
| Channel Features and Dimensions | Norman Creek (78 sites) | Moggill Creek (84 sites) | Bald Hills Creek (19 sites) | Nundah / Downfall Creek (78 sites) | Enoggera/ Breakfast Creek (124 sites) | Kedron Brook (82 sites) |
| Pool Depth (m) | 1.11 | 0.81 | 0.7 | 1.05 | 0.83 | 1.11 |
| Pool Width (m) | 6.5 | 9.7 | 12.2 | 18.3 | 7.7 | 9.8 |
| Non-pool Depth (m) | 0.84 | 0.72 | 0.33 | 0.98 | 0.78 | 0.83 |
| Non-pool Width(m) | 5.7 | 4.9 | 18 | 5.4 | 8.6 | 9.0 |
| Channel depth (m) | 3.3 | 3.3 | 1.8 | 3.1 | 2.0 | 2.9 |
| Channel Width (m) | 17.0 | 19.9 | 34.6 | 16.6 | 12.7 | 17.7 |
| Channel Width/Depth ratio | 6.0 | 8.4 | 38.5 | 6.2 | 9.9 | 7.7 |
| Channel Habitat Diversity | 12.8 | 32.6 | 5.2 | 21.9 | 31.8 | 20.0 |
| Bar size (% of bed) | 21.8 | 23.8 | 37.5 | 15.3 | 24.8 | 79.8 |
| Bankfull Width (m) | 15.0 | 18.6 | 18.4 | 14.4 | 23.1 | 17.9 |
| Bank Slope Width (m) | 4.6 | 3.6 | 1.8 | 3.1 | 3.1 | 4.1 |
| Stream side Zone Width (m) | 13.8 | 13.2 | 20.1 | 11.6 | 9.0 | 21.3 |
| Middle Zone Width(m) | 24.4 | 122.7 | 162.4 | 44.3 | 115.7 | 72.7 |
| Upper Zone Width(m) | 64.4 | 105.5 | 60.6 | 104.6 | 147.9 | 49.4 |
| Table 4. Comparison of Mean Stream Classification Parameters and Sediment Particle Sizes for the Six Pilot Catchments | ||||||
|---|---|---|---|---|---|---|
| Stream Classification | Norman Creek (78 sites) | Moggill Creek (84 sites) | Bald Hills Creek (19 sites) | Nundah / Downfall Creek (78 sites) | Enoggera/ Breakfast Creek (124 sites) | Kedron Brook (82 sites) |
| Average % natural for buffer | 57.6 | 96.2 | 66 | 81.6 | 70.5 | 85.9 |
| Bed Natural/Width Index | 3.9 | 5.9 | 5.7 | 5.7 | 4.8 | 5.3 |
| Bank Slope Natural/Width Index | 3.8 | 5.9 | 4.4 | 4.9 | 4.4 | 5.2 |
| Stream-side Natural/Width Index | 3.3 | 4.4 | 3.8 | 3.7 | 2.3 | 5.0 |
| Middle Zone Natural/Width Index | 2.7 | 3.9 | 3.4 | 2.9 | 3.7 | 4.5 |
| Upper Natural/Width Index | 3.7 | 4.8 | 3.4 | 3.9 | 3.0 | 5.0 |
| Access for Works | 42.5 | 44.6 | 47.6 | 42.3 | 45.0 | 46 |
| Average Buffer/Channel Width ratio | 4.5 | 37.4 | 20.5 | 9.25 | 5.0 | 11.1 |
| Suitability for Natural design | 29.3 | 73.7 | 65.4 | 62.5 | 72.6 | 56.6 |
| Constraints Index | 29.3 | 38.4 | 31.5 | 29.3 | 32.7 | 25.8 |
| Water Quality Problems Index | 6.4 | 2.2 | 3.6 | 13.4 | 7.8 | 15.5 |
| Pool sediment (mm) | 31.0 | 16.7 | 0.09 | 12.6 | 63.0 | 57.5 |
| Riffle sediments (mm) | 21.3 | 25.5 | 0.01 | 31.2 | 78.2 | 43.5 |
| Run Sediments (mm) | 102.0 | 10.8 | 0.12 | 48.0 | 40.7 | 57.8 |
| Glide Sediments (mm) | 84.1 | 16.8 | 0.18 | 79.4 | 44.3 | 0.09 |
| Cascade Sed. (mm) | 316 | 316 | 0.01 | none | 213.4 | 247 |
| Table 5. Comparison of Bank, Bed and Channel Habitat Types for the Six Pilot Catchments | ||||||
|---|---|---|---|---|---|---|
| Bank, Bed and Channel Habitat Types | Norman Creek (78 sites) | Moggill Creek (84 sites) | Bald Hills Creek (19 sites) | Nundah / Downfall Creek (75 sites) | Enoggera / Breakfast Creek (124 sites) | Kedron Brook (82 sites) |
| Bank Shape | Concave Convex Wide bench | Concave Convex Stepped | Concave Convex Low bench | Concave Convex Stepped | Concave Convex Stepped | Concave Convex Wide bench |
| Bank Slope | Steep Moderate Low | Low Moderate Steep | Moderate Steep Flat | Moderate Steep Low | Steep Moderate Vertical | Moderate Steep Low |
| Cause of Bank Instability | Ford Runoff Clearing | Flow Runoff Ford | Flow/wave Clearing Runoff | Flow/wave Clearing Runoff | Runoff Ford/road People | Tracks Runoff Ford |
| Bank Protection | Concrete Rock wall | Fencing Trees Rock wall | Trees Concrete Rip rap | Trees Concrete Rip rap | Trees Rock wall Concrete | Concrete Rock wall Rip rap |
| Location of Instability | Irregular All along | Irregular All along | Bends All along | All along Bends | Bends Irregular | |
| Causes of Bed Instability | Bank Erosion Bed deepening | Bank erosion Grazing | Bank erosion Bed deepening | Bank erosion Bed deepening | Bank erosion Channels | |
| Bed Controls | Bridge Installed Rubble | Fallen trees Bridge Rocks | Vegetation Fallen trees Rubble | Vegetation Trees Rubble | Rock outcrops Rubble | Ford Rubble |
| Bar Location | Point Alternate Island | Infilled Island Vegetation | Point Alternate Island | Point Alternate Obstruct. | Point Alternate Vegetation | |
| Channel Habitat Types (mean % of reach length)* |
| |||||
| Pools | 62 (40%) | 64 (95%) | 91 (74%) | 58 (65%) | 54 (65%) | 50 (50%) |
| Riffle | 35 (32%) | 31 (71%) | 38 (16%) | 31 (24%) | 15 (42%) | 40 (45%) |
| Run | 90 (56%) | 39 (28%) | 86 (32%) | 67 (59%) | 71 (49%) | 85 (55%) |
| Cascade | 5 (3%) | 6 (9%) | 0 | 4 (7%) | 25 (28%) | 15 (5%) |
| Rapid | 20 (1%) | 25 (19%) | 0 | 0 | 37 (21%) | 20 (2%) |
| * Figures presented show the average percent of reach length for the habitat type when present and the percentage of sites (in brackets) with the type present. | ||||||
| Table 6. Riparian Condition Ratings, Zone Dimensions and Mean percent cover and Percentage Occurrence of vegetation types at Sites the Six Pilot Catchments. | ||||||
|---|---|---|---|---|---|---|
| Norman Creek (78 sites) | Moggill Creek (84 sites) | Bald Hills Creek (19 sites) | Nundah/ Downfall Creek (78 sites) | Enoggera/ Breakfast Creek (124 sites) | Kedron Brook (82 sites) | |
| Riparian Vegetation Rating | 12 | 22 | 10 | 16 | 25 | 24 |
| Original Riparian Zone width(m) | 50.6 | 52.9 | 75.9 | 47 | 45.2 | 58.4 |
| Remnant Riparian Zone width (m) | 10.6 | 15.4 | 10.1 | 13.7 | 17.5 | 16.6 |
| Tall tree (>30m) | 38 (32%) | 21 (31%) | 0 | 25 (86%) | 25 (57%) | 29 (13%) |
| Medium tree (10-30m) | 30 (42%) | 55 (92%) | 25 (55%) | 30 (64%) | 41 (85%) | 41 (73%) |
| Small tree (<10m) | 21 (58%) | 28 (86%) | 16 (66%) | 18 (71%) | 27 (75%) | 26 (77%) |
| Shrubs (>2m) | 17 (29%) | 21 (20%) | 11 (13%) | 11 (31%) | 27 (48%) | 20 (24%) |
| Shrubs (<2m) | 16 (35%) | 29 (57%) | 14 (13%) | 9 (24%) | 26 (48%) | 33 (50%) |
| Rushes | 20 (35%) | 12 (61%) | 15 (85%) | 19 (77%) | 12 (61%) | 10 (50%) |
| Herbs | 17 (33%) | 31 (94%) | 16 (97%) | 24 (90%) | 37 (81%) | 24 (78%) |
| Grasses | 64 (98%) | 53 (97%) | 76 (100%) | 83 (96%) | 59 (89%) | 67 (90%) |
| Aquatic Vegetation | ||||||
| Submerged | 28 (31%) | 23 (50%) | 40 (26%) | 29 (57%) | 27 (53%) | 35 (55%) |
| Floating | 15 (1%) | 40 (4%) | 20 (3%) | 5 (1%) | 16 (4%) | 16 (16%) |
| Emergent | 49 (71%) | 27 (77%) | 62 (50%) | 50 (82%) | 22 (47%) | 24 (54%) |
| Table 7. Aquatic Habitat Condition Ratings and Mean percent cover and Percentage Occurrence of instream and Bank cover types at Sites the Six Pilot Catchments. | ||||||
|---|---|---|---|---|---|---|
| Aquatic Habitat | Norman Creek (78 sites) | Moggill Creek (84 sites) | Bald Hills Creek (19 sites) | Nundah / Downfall Creek (78 sites) | Enoggera/ Breakfast Creek (124 sites) | Kedron Brook (82 sites) |
| Condition Rating | 45 | 53 | 46 | 38 | 58 | 39 |
| Instream Cover | ||||||
| Logs | 8 (10%) | 12 (21%) | 0 | 4 (15%) | 11 (48%) | 9 (30%) |
| Log jams | 0 | 16 (3%) | 0 | 5 (1%) | 8 (22%) | 12 (9%) |
| Branches | 13 (23%) | 16 (88%) | 10 (36%) | 6 (20%) | 8 (60%) | 6 (13%) |
| Branch piles | 10 (11%) | 15 (67%) | 5 (50%) | 4 (13%) | 10 (55%) | 5 (18% |
| Algae | 25 (14%) | 26 (33%) | 22 (50%) | 23 (63%) | 32 (58%) | 25 (25%) |
| Submerged Vegetation | 43 (10%) | 22 (18%) | 64 (10%) | 28 (10%) | 23 (6%) | 27 (28%) |
| Emergent Vegetation | 43 (35%) | 16 (9%) | 68 (58%) | 36 (39%) | 26 (11%) | 20 (26%) |
| Tree Roots | 12 (18%) | 8 (6%) | 69 (5%) | 9 (11%) | 10 (30%) | 8 (11%) |
| Rock Surfaces | 37 (11%) | 42 (43%) | 0 | 27 (14%) | 49 (74%) | 43 (32%) |
| Pools > 1m deep | 28 (9%) | 40 (21%) | 0 | 40 (17%) | 51 (16%) | 50 (13%) |
| Instream Cover Rating (%) | 50 | 68 | 63 | 54 | 78 | 51 |
| Bank Cover | ||||||
| Canopy Cover | 42 (70%) | 51 (94%) | 26 (53%) | 23 (63%) | 43 (93%) | 42 (57%) |
| Vegetation Overhang | 58 (72%) | 52 (94%) | 69 (71%) | 59 (67%) | 43 (83%) | 38 (51%) |
| Root Overhang | 14 (16%) | 14 (18%) | 0 | 20 (10%) | 20 (22%) | 16 (50%) |
| Bank Overhang | 10 (6%) | 19 (20%) | 0 | 10 (8%) | 13 (17%) | 8 (10%) |
The pilot studies were very successful. The opportunity to conduct full and complete surveys of six catchments provided a much better evaluation of the methods and enhanced the modification of the system in response to the feed-back from the survey participants. It also meant that the data and reports provided real and complete information about the six catchments. The studies became full evaluation exercises rather than pilot studies.
The pilot studies clearly showed that a wide range of non-expert personnel, who are trained to conduct the surveys and to run the software system, could conduct the surveys. The community group volunteers performed well during the surveys and their local knowledge was invaluable for selecting the sites and defining the boundaries for the 'homogeneous stream sections'.
The surveys were generally completed within the allotted time frame and most teams accomplished the expected rate of survey of eight sites per day. Supplementary surveys by two teams over several days were conducted to complete the sites omitted and to cover the 'bonus' sites and sites missed due to wet-weather and equipment problems.
The resources needed for the expansion of the pilot to six major surveys were inadequate in terms of the staff needed to complete the data entry phase of the project. This caused unfortunate delays in data processing and production of data summaries for BCC. A smaller pilot would have been completed sooner. Likewise the data entry and analysis would have been completed sooner if the software system had been completed, rather than under development. Nevertheless these disadvantages were offset by the benefits of conducting six complete evaluation surveys and produced real data and output.
Adequate time and experienced personnel need to be made available for the data entry and validation. The staff used should be familiar with the survey methods or better still to have been involved in the surveys. Many data errors can be eliminated at this stage. Normally about 15-20 sites can be entered per day. Once the data has been entered and verified the data analyses and outputs can be quickly produced, usually within several hours.
The data summaries included in this report provide a comparison between the catchments in terms of condition, type and various classification parameters. These data are presented mostly for demonstration purposes. However, some discussion of the results is warranted to show how the data can be used and interpreted. The standard set of condition categories and descriptors are used in these data summaries are:
| Condition Category | Stability Descriptor | Habitat Descriptor |
| 0-20% | 'Highly Unstable' | 'Very Poor' |
| 21-40% | 'Very Unstable' | 'Poor' |
| 41-60% | 'Moderately Unstable | 'Moderate' |
| 61-80% | 'Partially Unstable' | 'Good' |
| 81-100% | 'Stable '' | 'Very Good' |
Anderson (1999a) provides an explanation of what these terms mean for each of the attributes.
The overall condition ratings provide a general summary of the overall condition of the habitat in the channel and the riparian vegetation (Anderson 1999a). This rating is produced as an average of the condition ratings for:
Double weightings are applied for riparian vegetation and aquatic habitat.
Mean overall condition ratings for the sites in each of the pilot catchments ranged from a minimum of 37% in Bald Hills Creek to 50% in Enoggera Creek (Table 2). Most of the sections were rated as having 'Moderate' (41-60%) to 'Poor' (21-40%) overall condition (see maps in Appendix 4). Enoggera and Nundah Creek had 18% and 37% of their stream length rated in 'Good' overall condition (61-80%).
The low overall condition ratings were mostly attributed to the poor condition of the riparian vegetation and aquatic habitat. About 50% of the length of stream in all catchments except Enoggera Creek had 'Very Poor' riparian vegetation (0-20%). Remnant zones were generally very narrow. More than 50% of the length of stream in all the pilot catchments had remnant riparian zone width less than 20m wide and 25% less than 10m wide.
More than 50% of the length of stream in each catchment had aquatic habitat ratings of 'Moderate' to 'Good'. More than 35% of the length of stream in Nundah and Norman Creeks and Kedron Brook had 'Poor' to Very Poor' aquatic habitat ratings. Bald Hills, Enoggera and Moggill Creek had more than 70% of the stream length rated as 'Moderate' or 'Poor' for aquatic habitat. These low scores contributed to the low overall condition rating for most of the stream length.
Bed and Bank ratings were generally high in all six catchments except for Moggill and Bald Hills Creeks.
The 'Skeleton' maps show that most of the sections rated in the range from 26-40% (coloured green) and 41-60% (coloured black) for overall condition (Appendix 4). Most of the sections with higher overall ratings occurred in the headwaters of the catchments or in the mangrove-lined sections near the mouth of the rivers. Sections with higher overall ratings were found in the central area of the Enoggera Creek catchment, downstream of the reservoir as well as in the upper parts of the Enoggera Creek and its tributaries (Appendix 4).
The overall condition rating provides an effective audit of the condition of the streams in the catchments for comparative purposes. It also highlights problem areas and potential conservation and reference sites. The reasons for the low or high overall condition ratings can be resolved by examining the scores of the component ratings and by examining the raw survey data using the software system.
The mean riparian vegetation condition ratings for the sites were generally low and ranged from about 25% in Enoggera Creek and Kedron Brook to less than 10% in Bald Hills and Numdah Creeks (Table 2). These low ratings were generally attributed to the loss of most of the original riparian zone width and low tree and shrub cover. Mean remnant riparian zone widths ranged from about 18m in Enoggera Creek and Kedron Brook, to only 5m in Nundah Creek (Table 2). There was considerable variation in riparian vegetation ratings in different parts of the catchment. More than 20% of the length of stream in Nundah Creek, Kedron Brook, and Enoggera Creek was rated as 'Good' to 'Very Good' (rating >60%) (Appendix 2). 'Very Poor' riparian vegetation condition was found for 45% of the stream length in Enoggera Creek, 83% in Norman Creek and 67% in Bald Hills Creek. Most of the sections with good remnant riparian vegetation were found in the less developed headwater streams and tributaries (Appendix 4), but there were pockets of good vegetation in the middle and lowland reaches.
Trees and shrubs occurred at more sites and had higher mean cover ratings in Enoggera and Moggill Creeks, and Kedron Brook. Trees were less prevalent and dense in Bald Hills and Norman Creeks (Table 6).
Submerged aquatic vegetation occurred at about 50% of the sites in Enoggera and Moggill Creeks and Kedron Brook, but only 25-30% of the sites in Norman and Bald Hills Creek (Table 6). Floating vegetation was rare. Emergent vegetation was found at most sites and average percent covers ranged from 22% in Enoggera Creek to 62% in Bald Hills Creek.
Mean aquatic habitat condition ratings for the sites ranged from 38% and 39% in Nundah Creek and Kedron Brook to 53% and 58% in Moggill and Enoggera Creeks (Table 7). Instream cover ratings were generally higher than the bank ratings, reflecting the loss of riparian vegetation. Mean percentages of bank length with canopy cover present were generally low ranging from 23% in Nundah Creek to 51% in Moggill Creek (Table 7). Canopy cover was absent from about half of the sites in Bald Hills Creek and Kedron Brook (Table 7). Submerged and emergent vegetation provided most of the aquatic habitat as organic debris cover. Logs and branches were uncommon and covered and average of 10% of the bed in Moggill and Enoggera Creeks (Table 7).
'Good' to 'Very Good' Aquatic habitat ratings (>60%) were found for about 30-40% of the stream length in Norman, Moggill, Enoggera, and Bald Hills Creeks, but only about 20% in Kedron Brook and Nundah Creeks (Appendix 2). The more highly rated sections in each catchment provide reference sites for rehabilitation initiatives and also represent potential conservation sites. 'Moderate' to 'Good' aquatic habitat ratings were found in some sections where the banks and buffer areas had been highly modified, but the absence of canopy cover and low riparian vegetation cover restrict the potential of these sites for aquatic habitat.
Most of the sections in Moggill and Enoggera Creek showed poor to moderate aquatic habitat ratings being either 21-40% (colored black) or 41-60% (colored green) (Appendix 4). There were relatively few sections in very good or very poor condition. About half of the sections in Kedron Brook were rated as 'Very Poor' (0-20%) (colored red) or poor 21-40% (colored dark blue). These poorer condition sections were found throughout the catchment. Norman Creek, Nundah and Bald Hills Creek showed highly variable aquatic habitat ratings (Appendix 4).
Bank stability was generally high in the pilot catchments. Average site ratings ranged from about 57% in Moggill Creek to 90% in Kedron Brook, though these figures include highly modified channels with concrete linings or modified and mown floodplain areas. Most of the banks were concave with moderate to steep slope (Table 5). About 23%, 47% and 30% of the length of stream in Nundah, Bald Hills and Moggill Creeks, respectively, had banks rated as 'Unstable'. Enoggera and Norman Creeks had less than 6% of bank length rated as 'Unstable'.
Mean bed and bar stability ratings for the sites were also generally high ranging from 48% in Kedron Brook to 81% in Norman Creek (Table 2). Most of the instability in the beds was linked with the presence of bars and bed deposits especially in Kedron Brook (Appendix 2) where about 42% of the bed was rated as unstable.
These results highlight the different approaches needed to manage the six pilot catchments in terms of the problems and processes affecting bed and bank stability and habitat condition.
There was considerable variation in the prevalence of channel habitat types and sediment particle sizes in the pilot catchments. Most sites surveyed in Moggill Creek had pools and riffles present with an average of 64% of the reach length as pool habitat. (Table 5). Pools occupied an average of more than 50% of the reach length for all the pilot catchments. Norman, Nundah and Kedron Brook had highest occurrence and percentage of run habitats, though, these included highly modified and channelised sections. Cascades and rapids were only prevalent in Enoggera and Moggill Creeks.
Bald Hills Creek and Nundah Creek had predominantly silt/clay sediments in pools and runs (Appendix 2). Mean particle sizes in Moggill Creek were predominantly in the medium to large gravel range in pools and riffle/runs. In Norman Creek and Kedron Brook sand and silt/clay predominated in the pools and gravel in the riffle/runs (Appendix 2). Mean particle sizes in the cobble range were found at more than 20% and 7% of the length of pools and riffle/cascades in Enoggera Creek and Moggill Creek, respectively. Cobbles were absent from most of the length of stream in the other pilot catchments.
Once again these differences illustrate the different approach required for the various pilot catchments.
Most pools were less than 1m deep, at the water mark. About 27%, 15% and 11% of the length of pools in Nundah Creek, Kedron Brook, Enoggera Creek, respectively, were deeper than 1.5m. Nundah and Enoggera Creek also had the widest pools (Appendix 2). Channel width to depth ratios was highest in Bald Hills, Enoggera and Moggill Creeks. Norman and Nundah Creeks had the narrowest 'Stream side' and 'Middle Buffer' zone widths (Table 3).
Average buffer naturalness and the channel and buffer modification index was useful for summarizing the extent of modification of the bed banks and buffer zones. The modification index was produced as follows (see Anderson 1999a for details):
A stream section with natural bed and banks but with a highly modified stream side zone that is paved or covered with mown grass or parkland will have a index of 66333, or 66444. Even if this section had a natural streamside zone the index may be 66533 or 66433 if the width of the zone is inadequate in relation to the bankfull width (see Anderson 1999a).
Table 8 provides an explanation for the categories used to display the data for this index. The categories chosen reflect the key boundaries in ranking the sections in terms of the extent of modification to the bed, bank and buffer zones. The higher the value of the index, as a number, the higher the overall naturalness and potential value of the stream section. This index reflects the relative value of the different zones. The bed condition is ranked higher than the bank, and this in turn ranked higher than for the streamside zone. The buffer zones are effectively ranked in terms of their proximity to the stream.
This index can also be used to define the capability of the stream section for rehabilitation. For example, the overall value of the section as defined by the index will be more heavily weighted by the modifications to the bed, banks and streamside zones. The net benefits of improving the riparian vegetation by rehabilitation or restoration will be set by the naturalness of the bed and bank slope. If the bed is highly modified, and the banks are natural and the buffer zones are at '222' the net maximum shift in code will be from an index of 16222 to 16622. If the bed was originally natural the benefit will be to shift the index from 66222 to 66622. It also follows that maximum benefits would be obtained from restoring the riparian vegetation in sections that had natural bed. The condition of the bed would also be likely to be improved by regenerating the riparian vegetation by increasing canopy cover and bank vegetation cover and indirectly by restricting sediment transport to the stream and by providing branches and organic debris to the stream. These benefits will be greatly restricted if the bed is highly modified. Likewise the benefit will be restricted if the zone is too narrow to function effectively. Restoring vegetation in narrow and highly restricted riparian zone will produce a restricted benefit.
Given the ability to choose where the restoration will occur, the simple principle is to choose the section that has the highest index score treated as a number for the zones closer to the stream. Usually this will simply mean choosing sections with the higher index as a number. For example bank restoration should be targeted at sections above 40,000 which have beds that are more than 70% natural. There may also be benefits in choosing sections with wider and adequate buffer zones to maximize the potential benefit.
The average channel and buffer naturalness score is produced by weighting the scores for the components. Higher weightings are applied for the zones closer to the stream (Anderson 1999a).
These indices provide a way of comparing the sections in the pilot catchments in terms of the modifications present and their potential for rehabilitation.
About 90% of the stream length in Moggill Creek and 79% in Enoggera Creek were classified as natural with no major modifications (Table 9). Kedron Brook and Norman Creek had more modified channels with only 36% and 46% of the channel length being classified as natural. Bald Hills and Nundah Creeks showed intermediate percentage of natural channel (54%).
The average buffer naturalness score was also very high in Moggill Creek (96%; Table.4), and 98% of the stream length had buffer naturalness scores of over 80% (Table 9). Average buffer naturalness scores were lowest in Norman Creek, with about half of the stream length having a naturalness score of less than 60% (Table 9). The average naturalness scores for the other catchments were between 66 and 86% (Table 4).
| Table 8. Explanation for Channel Bank and Buffer Zone Modification Index Categories. | |||||
|---|---|---|---|---|---|
| Category | Zone modification and Width Adequacy | ||||
| Bed | Bank | Streamside | Middle | Upper | |
| 10000-50000 | modified | x | x | x | x |
| 50000-60000 | slightly modified / narrow | x | x | x | x |
| 60001-66000 | natural | modified | x | x | x |
| 66001-66500 | natural | natural | modified | x | x |
| 66501-66600 | natural | natural | Slightly modified / narrow | x | x |
| 66601-66660 | natural | natural | natural | modified | x |
| 66661-66664 | natural | natural | natural | natural | modified |
| >66664 | natural | natural | natural | natural | natural |
Table 9 provides a summary of the channel, bank and buffer zone modification indices for the pilot catchments. The highest percentage of the stream length with lower index scores was found in Norman, Enoggera and Nundah Creeks, and Kedron Brook. More than 20% of the stream length in these catchments had indices less than 60000 (Table 9). On the other hand more than 50% of the stream length in Moggill, Bald Hills and Kedron Brook had indices greater than 66500 (natural bed, banks and streamside zones). These latter catchments have natural bed and banks with less channelisation and concrete artificial channels than the other catchments. About 16% of the stream length in Kedron Brook and 7% in Enoggera Creek had natural beds, banks, and buffer zones (indexes above 66664; Table 9).
The indices for Enoggera Creek appeared to be lower than expected given the relatively high buffer naturalness index of about 70%. However 22% of the length of stream had buffer to bankfull widths less than 2 (Table 4), and the average streamside zone index was 2.3 (Table 4). The combination of narrow buffer zones and highly modified streamside zone has reduced the channel and buffer modification index. Moggill Creek also had the lowest percentage of stream length with narrow buffer widths.
The distribution of the index categories shown in Table 9 for the six pilot catchments is shown in (Appendix 4). These plots are useful for identifying modified sections, and for setting management and rehabilitation priorities in relation to the buffer widths available. In Norman Creek the highest index scores were found in Sandy Creek, in the headwaters of Ekibin Creek and in the lower reaches near the mouth of Norman Creek. (Appendix 4). In Nundah Creek the more natural sites were in the lower reaches upstream and downstream of Zillman Waterholes junction, and also in this major tributary. In Enoggera Creek the highest indices were found in Ithaca Creek and an unnamed tributary downstream of Enoggera Reservoir. The natural stream sections, upstream of the reservoir, also had high ratings, as was expected. Narrow buffer zones reduced that score in some of these sections. The downstream reaches generally had modified bed and banks.
Most of the bed and banks were natural in Moggill Creek and indices were mostly above 66000 (Appendix 4). Sections with natural streamside zones were found in the unnamed tributaries. The highest indices in Kedron Brook occurred in isolated pockets in the headwater streams and unnamed tributaries, but most sections had indices less than 666000 reflecting the modified buffer zones and channelised sections in the middle of the catchment. Most of these areas had indices from 60000 to 66000.
| Table 9. Channel, Bank and Buffer Zone Modifications in Six Pilot Catchments as percentages of stream length surveyed. | ||||||
|---|---|---|---|---|---|---|
| Attribute | Norman Creek | Moggill Creek | Bald Hills Creek | Nundah / Downfall Creek | Enoggera / Breakfast Creek | Kedron Brook |
| Length of stream surveyed (km) | 32 | 42 | 16 | 33 | 80 | 64 |
| 'Natural Channel Shape | 46 | 95 | 57 | 57 | 58 | 37 |
| Natural Unmodified Channel | 46 | 96 | 54 | 54 | 79 | 36 |
| Average Buffer Naturalness | ||||||
| >80% | 48 | 98 | 87 | 66 | 66 | 82 |
| 61-80% | 6 | 0 | 0 | 16 | 18 | 10 |
| <61% | 46 | 2 | 13 | 18 | 15 | 8 |
| Buffer Naturalness and Adequate Width Index | ||||||
| >66664 | 0 | 0 | 2 | 2 | 7 | 16 |
| 66661-66664 | 0 | 0 | 0 | 2 | 0 | 0 |
| 66601-66660 | 4 | 5 | 7 | 4 | 0 | 1 |
| 66501-66600 | 34 | 62 | 58 | 20 | 16 | 50 |
| 66001-66500 | 9 | 21 | 11 | 20 | 27 | 2 |
| 60001-66000 | 6 | 2 | 2 | 19 | 5 | 6 |
| 50000-60000 | 2 | 6 | 2 | 16 | 30 | 11 |
| 10000-50000 | 40 | 0 | 3 | 7 | 15 | 9 |
| Suit Natural Design | ||||||
| >80% | 37 | 34 | 39 | 19 | 50 | 9 |
| 61-80% | 45 | 61 | 49 | 49 | 39 | 38 |
| <61% | 18 | 5 | 12 | 31 | 12 | 53 |
| Constraints Score | ||||||
| >80% | 1 | 0 | 0 | 0 | 0 | 0 |
| 61-80% | 1 | 0 | 0 | 3 | 4 | 1 |
| 41-60% | 24 | 0 | 18 | 18 | 14 | 8 |
| 21-40% | 41 | 4 | 40 | 44 | 58 | 40 |
| 0-20% | 33 | 96 | 42 | 35 | 23 | 51 |
| Average ratio of buffer to bankfull width | ||||||
| <2.0 | 15 | 5 | 27 | 32 | 22 | 28 |
| 2-3.0 | 19 | 3 | 14 | 2 | 7 | 9 |
| 3.1-5.0 | 30 | 36 | 16 | 7 | 4 | 13 |
| 5.1-7.0 | 23 | 29 | 4 | 12 | 7 | 15 |
| 7.0-10.0 | 6 | 7 | 0 | 3 | 7 | 5 |
| >10.0 | 7 | 20 | 38 | 44 | 52 | 30 |
The highest suitability for natural design index scores occurred in Moggill Creek with 95% of the stream length rated as >60% (Table 9). The lowest scores were found in Kedron Brook and Nundah Creek where 53% and 31% of the stream length, respectively had scores less than 61%. The constraints index was also lowest in Moggill Creek, Kedron Brook, Bald Hills Creek, and Nundah Creek with 96%, 51%, 42% and 35%, respectively, of the stream length with scores less than 20% (Table 9).
The formula function for the 'Skeleton' map package allows the selection and ranking of sites for various combinations of the classification indices using formulae (Anderson 1999a). The first exercise was to select and rank the sections in terms their potential for rehabilitation of aquatic habitat using artificial riffles or natural design principles. Only sections with natural bed were selected. The suitability ranking was produced by using the natural design suitability rating as a positive weighting. The aquatic habitat, constraints index and channel habitat diversity indices as negative weightings. This increased the score for sections that had low habitat diversity, low habitat diversity and few if any constraints for rehabilitation. The first output shown in Appendix 6, was generated by applying a formula with the following attributes and weightings:
| Attribute | Weight | Selection Test |
| Suitability for Natural Design | 200 | |
| Aquatic Habitat Condition | -50 | |
| Constraints Index | -50 | |
| Channel Diversity Score | -50 | |
| Channel Modification Index | >40000 (bed >70% natural) |
The output for this test for the Norman Creek catchment is shown in Appendix 6. Most of the sections were eliminated (colored grey) because the bed was modified. The selected sections were rated from 4-6 (colored black, yellow and blue respectively). Sections 3384 and 3339 in Sandy Creek received the highest ranking (Appendix 6.). These sections are highlighted as having high potential for rehabilitating aquatic habitats by creating riffles and pool areas or by invoking natural design criteria and recreating pool and riffle sequences. The outputs are rankings of potential suitability. The results are only indicative of a potential suitability. There is a need for more detailed investigation of these sections to make the final selection. However the selection of suitable sections and the creation of relative suitability scores is very helpful making these decisions. Other formulae and different weighting may provide more meaningful results.
The purpose of this demonstration was to rank the sections in terms of their potential as good fish habitat, either for conservation or for re-establishing fish population by stocking or re-allocation. The key criteria selected were aquatic habitat ratings, channel habitat diversity and riparian vegetation condition. Sections were only considered if the bed was natural and pools were present and deeper than 0.5m (at the water mark).
The score for these criteria was produced by using the following attributes and weightings:
| Attribute used for rating | Weight | Selection Test |
| Aquatic Habitat Condition | 100 | |
| Channel Diversity Score | 100 | |
| Riparian Vegetation Condition | 100 | |
| Selection Attributes | ||
| Pool depth > 0.5m @ wm | >5 (value is multiplied by 10) | |
| Channel Modification Index | >40000 (bed >70% natural) | |
This formula selects sections that have natural bed (more than 70% natural), and a high rating for aquatic habitat, riparian vegetation and a diversity of channel habitat types. The use of pool depth eliminated sites with no pool present or with shallow and less permanent pools. The output for this test for the Norman Creek catchment is shown in Appendix 6. Relatively few sections met the criteria. The sections with highest potential were located in Norman Creek upstream of the Mott Creek confluence, and in the headwaters of the catchment. Section 3375, an unnamed tributary near the mouth of Norman Creek was also highly ranked. The rankings provide a strategy for further investigation of these sections.
The purpose of this demonstration was to find sections where trees planting would provide the greatest benefit to the stream and buffer zones. The key criteria selected were naturalness of the bed, banks and buffer zones, stable banks and a minimum remnant riparian zone width of 10m. Sections with low riparian vegetation condition were scaled higher, by using a negative rating for riparian vegetation. The following attributes and weightings were used to generated this rating:
| Attribute used for rating | Weight | Selection Test |
| Average naturalness score | 100 | |
| Bank Stability | 100 | |
| Riparian Vegetation Condition | -50 | |
| Selection Attributes | ||
| Remnant riparian zone width >10 m | >10 | |
The highest ranking sections for these criteria were in Sandy Creek (sections 3328,3329 and 3384) and in the upper reaches of Ekibin Creek (3318, 3357 and 3394).
The purpose of this demonstration was to rank the sections in terms of their potential as macroinvertebrate reference sites. The key criteria used were aquatic habitat condition, channel habitat diversity rating, riparian vegetation condition, naturalness of the bed, banks and buffer zones. Sections were selected that had riffle areas present, and with riffles sediment particle size greater than 2mm (gravel).
The following attributes and weightings were used to generated this rating:
| Attribute used for rating | Weight | Selection Test |
| Aquatic Habitat Condition | 100 | |
| Channel Diversity Index | 100 | |
| Riparian Vegetation Condition | 100 | |
| Average naturalness score | 100 | |
| Selection Attribute | ||
| Riffle sediment particle size> 2mm | >2 | |
The highest ranking sections for these criteria were found in Mott Creek and Norman Creek immediately upstream and downstream from the junction. High-ranking sections were also found in the tributary sections in the headwaters of Norman Creek and Ekibin Creek (Appendix 6).
The purpose of this demonstration was to rank the sections in terms of their potential as conservation or reference sites for riparian vegetation The key criteria used were riparian vegetation condition, conservation rating, bank stability rating, and naturalness of the bed, banks and buffer zones. Sections with highly modified bed were eliminated. The following attributes and weightings were used to generated this rating:
| Attribute used for Rating | Weight | Selection Test |
| Riparian Vegetation Condition | 100 | |
| Conservation Rating | 100 | |
| Bank Stability Rating | 100 | |
| Average naturalness score | 100 | |
| Selection Attribute | ||
| Channel Modification Index | >40000 (bed >70% natural) | |
The highest ranking sections for these criteria were found in Mott Creek and the sections immediately upstream and downstream from it, and in the tributary sections in the upper parts of Norman Creek and Ekibin Creek (Appendix 6).
The various plots for the selected raw data attributes are shown in Appendix 7. These plots illustrate how the raw data for the sites can be used for management decision. The plots highlight sections with unstable aggrading beds and eroding banks. They also show the location of sites heavy weed infestation and various channel habitat and vegetation attributes. Most of the raw data attributes can be simply and easily displayed on screen or as printed output for all of the pilot catchments.
Table 10 provides a general summary of the time and personnel needed for conducting the surveys, undertaking the analyses and generating the reports and data summaries. The summaries have been prepared for a catchment with 200 sites. The pilot surveys suggest that 200 sites would cover about 150 km of urban stream and about 400 sq. km of catchment area. These sites would cover the surveys for Kedron Brook and Enoggera Creek, or Norman, Moggill and Bald Hills Creeks. However, estimates of the time required are also provided in terms of the sites processed for each person-day. This allows the estimates to be geared up for larger catchments or changed to survey multiple catchments.
It is suggested that a full time coordinator would be required for 12 weeks. This person would organise the surveys, assist in the surveys themselves and undertake or supervise the data entry and analysis. A part-time administrator would be required for about eight weeks to organise the appointment of staff and the planning and training workshops. This person would also communicate with the funding organization and the client for the project and would have input to the final reports. Community groups would be involved in the planning workshops for each catchment, the reconnaissance surveys and also in preparing the final reports. It is suggested that three teams of two people would be required for the surveys, including volunteers from the community groups. Data entry and analysis would require two weeks for a single person. Defining the section boundaries and sub-catchment polygons for digitizing for the GIS would require about one week. This resource summary does not include the development of the GIS, but only the preparation of the information. One week has been allocated for preparing he data summaries and two weeks for preparing the draft report.
Table 10 Summary of the Estimated Time and Personnel Required for a 200 Site Catchment. | ||||||
|---|---|---|---|---|---|---|
| Time Table | Tasks | Personnel: COOR coordinator; ADM administrator; COM community group; SUR survey team | Completion rate (Sites per day) | |||
| COOR (1) | ADM (1) | COM (1) | SURV (5) | |||
| Week 1 | Preliminary planning and team selection | |||||
| Week 2 | Gathering existing data Preliminary Planning Workshop | Extra days for larger or additional catchments | ||||
| Week 2 | Staff Appointment and Organization of Equipment | One vehicle and set of gear require for each team of two | ||||
| Week 3 | Reconnaissance Surveys | 20 sites a day for a each team of two (community groups involved) | ||||
| Week 4 | Reconnaissance Surveys | |||||
| Week 5 | 2-Day Training Workshop | Administrator organizes workshop & supplementary training | ||||
| 4 days of survey (one training day) | 8 sites a day for each team of two | |||||
| Week 6 | 5 days of surveys | 8 sites a day for each team of two | ||||
| Week 7 | Data Entry and Verification | 20 sites a day for data entry | ||||
| Week 8 | Data Entry and Verification | |||||
| Week 9 | Mapping of Sections and Catchment polygons 'Skeleton' Maps GIS Mapping | More time required for larger catchments and multiple catchments. | ||||
| Week 10 | Data Analysis and production of data summaries and reports Transfer of data to GIS | Data analysis only requires about an hour and the data summaries can be produced quickly | ||||
| Week 11 | Preparation of Draft Reports Training for use of Software system | Training workshops are needed for the software package | ||||
| Week 12 | Preparation of Draft reports Training for use of Software system | |||||
These summaries have been provided for:
The standard set of condition categories was used (0-20%, 21-40%, 41-60%, 61-80% and 81-100%).
 |
Appendix 1 Page 1 (original size graphic) |
(The remaining data sheets for this appendix are available as graphic and text images in the hard copy version - B Hall)
These summaries have been provided for:
The 'Urban' set of condition categories was used (0-10%, 11-25%, 26-40%, 41-60%, 61-100%). These summaries show the actual length stream classified in various ways, as well as the percentage of the total length of stream surveyed.
 |
Appendix 2 Page 1 (original size graphic) |
(The remaining data sheets for this appendix are available as graphic and text images in the hard copy version - B Hall)
Five report cards are provided for the first five planning units in each catchment, except for Bald Hills Creek, which only had two planning units. The report cards are only relevant for a small group of sites such as those in a section or planning unit. These summaries are listed in the following order:
 |
Appendix 3 Page 1 (original size graphic) |
(The remaining data sheets for this appendix are available as graphic and text images in the hard copy version - B Hall)
Skeleton Maps are provided for:
Four maps are produced for each catchment showing the following attributes in colour-coded categories:
 |
Appendix 4 Page 1 (original size graphic) |
(The remaining data sheets for this appendix are available as graphic and text images in the hard copy version - B Hall)
Six maps are provided for additional ratings in Norman Creek:
 |
Appendix 5 Page 1 (original size graphic) |
(The remaining data sheets for this appendix are available as graphic and text images in the hard copy version - B Hall)
Various formulae and selection criteria have been used to produce ratings for the following purposes:
 |
Appendix 6 Page 1 (original size graphic) |
(The remaining data sheets for this appendix are available as graphic and text images in the hard copy version - B Hall)
The package allows most of the raw data attributes collected during the surveys to be classified into rating categories and displayed on 'Skeleton Maps'. The attributes plotted were:
 |
Appendix 7 Page 1 (original size graphic) |
(The remaining data sheets for this appendix are available as graphic and text images in the hard copy version - B Hall)
Statistical Summary Reports have been provided for:
 |
Appendix 8 Page 1 (original size graphic) |
(The remaining data sheets for this appendix are available as graphic and text images in the hard copy version - B Hall)
Riversinfo Australia (au.riversinfo.org): Publication Information:
Current at December 2002
au.riversinfo.org has been developed to facilitate public access to scientific and technical information on Australia's Rivers. Much of this material was paid for using public funds and is therefore owned by the Australian people. Environment Australia have been instrumental in making available all the publications from:
The au.riversinfo.org Internet site is in-part supported by the Natural Heritage Trust. The Natural Heritage Trust focuses on five key environmental themes - land, vegetation, rivers, coasts and marine, and biodiversity. The programs of the Natural Heritage Trust play a major role in developing sustainable agriculture and natural resource management, as well as protecting our unique biodiversity through improved management and delivery of resources.
The au.riversinfo.org Internet site is in-part supported by Sydney Airport who have co-funded a Natural Heritage Trust supported project intended to develop Internet based environmental information integration and management systems. This will assist in developing an overall perspective on the demands and uses for environmental resources