Figure 1. Time series changes in relative value and utility of streams, channels, riparian zones and floodplain/valley flats with degradation, rehabilitation and restoration.
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.
Report No. 1. Development of the Classification System for Urban Streams
Dr. John R. Anderson
Centre for Coastal Management Southern Cross University
LWRRDC Occasional Paper 8/99 (Urban Subprogram Report No.1 1999)
Foreword NRHP, Publication Details & Acknowledgements
4. Application Of The Classification System For Management Of Urban Streams
6. Limitations Of The Classification System
7. Trial Of Methodology In Brisbane
List of Abbreviations Figures & Tables
Appendix 2. Condition Audit Summaries – Tabulated Output
Appendix 3 Condition Audit Summaries – 'Stack Diagrams'
Appendix 6 Statistical Summary
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 |
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).
Disclaimer:
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.
Publication data:
Basic Decision Support System for Management of Urban Streams - Development of the Classification System for Urban Streams', Anderson J R. National River Health Program, Urban Sub Program, Report No 1, LWRRDC Occasional Paper 8/99.
ISSN 1320-0992
ISBN 0 642 26758 8
Author:
Dr John R Anderson
Centre for Coastal Management
Southern Cross University
PO Box 157
Lismore NSW 2480
Telephone: 0266203009
Facsimile: 0266212669
Email: janders3@scu.edu.au
Managing Agencies
I acknowledge the assistance of Paul Mack and Tony Weber from the Brisbane City Council and Damien Madden from the Norman Creek Catchment Management Group who provided valuable feed-back during the development of the package. I would particularly like to thank Rhona McPhee who organised the pilot surveys and provided most of the organisation for the project and liaison with Brisbane City Council. I also thank the large number of community group members who provided tireless and enthusiastic support for various parts of the project.
Aims of the Study
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, capability for rehabilitation or enhancement, physical and environmental condition, and key constraints limiting restoration. The major features of this classification were:
| A multiple classification system was developed. |
The approach adopted was to develop a series of classifications rather than a single integrated classification system. A database system was preferred so that survey data and various derived data could be selected in various combinations to address specific issues. A hierarchical system allows the user to see how the derived ratings were produced and the raw data can be used to refine the decision process. The user has maximum flexibility and power to choose the data required for making decisions for managing streams.
|
Condition ratings for a wide range of habitat and other components are calculated as percentages of the original pristine value, function or utility that is retained. This provided a multi-faceted environmental audit. |
The condition of various habitats and stream attributes was assessed as percentage scores with 100% representing virtually pristine condition with maximum value and full function retained. Very low ratings represented almost complete loss of value or function. Formulae were used to generate absolute condition ratings, using a comprehensive set of objective data. The ratings can be scaled using local reference sites in very good or pristine condition to ensure the ratings are realistic in terms of local features. The absolute ratings and scaled ratings were both stored in the databases after being calculated by the programs.
Condition was assessed for the following components:
Additional information was collected on:
Various other derived ratings and summary attributes were produced:
|
Specific purpose surveys were required to consistently and completely assess entire catchments and all the waterways within the drainage network that had a permanent and defined channel. |
Consultation with various community and local government groups showed that specific-purpose surveys were required. Rapid survey techniques have been developed to gather data on the condition and modifications to urban streams and other data to classify the instream and riparian habitats and the channel morphology.
|
The sampling strategy was based on sub-dividing all the waterways in the drainage network into 'homogeneous stream sections' of varying length. One or more sites (=reaches) representative of the sections were selected and the data applied to the entire section. A local sub-catchment element was defined for each of the sections. The entire drainage network and catchment area are included for a project. |
All waterways in the drainage network were divided into 'homogeneous sections' of varying length. The waterways and buffer zones within these sections had similar condition, habitat types present, channel morphology and similar modifications to the channel, banks and buffer zones. Nodes in the drainage network and other attributes were used to define various boundaries for these sections. The sub-sectioning continued until the test of 'homogeneity' was satisfied at the scale, emphasis and detail required for the study. The sections were first defined as a mapping exercise, and then checked and further refined using reconnaissance surveys. One or more representative reaches (sites) were selected during the reconnaissance for detailed survey. The data collected from the sites was then applied to the entire section. The lengths of the sections were measured. Reports on the condition of the waterways could then be expressed as a total length (or percentage) of waterway classified in particular ways. A quantitative audit of waterway condition and distribution of key features could be produced for catchment in terms of stream length classified into various categories. For example, surveys may show that 80% of the length of banks in the waterway (say 27km) were highly unstable, either severely eroding or slumping (bank condition less than 20%). The surveys may also show that 25% of the stream length had Aquatic Habitat in 'Very Good' condition (ratings >80%), but also 45% in 'Very Poor' condition (ratings <20%). Other software is designed to show the distribution of various classified sections in user-defined condition categories. Various classified habitat types can also be mapped using an internal mapping system or through downloading the section data to a GIS.
|
A total of 13 datasheets designed for rapid surveys of urban corridors (one site per hour) by 'non-expert' personnel who undertake a 2-day training exercise. |
A set of 13 datasheets have been carefully designed using graphic techniques to eliminate coding sheets and enable the surveys to be conducted by non-experts, who are trained for the task. Local Government employees, university students and community group volunteers were used for the pilot studies. The surveys were designed for a team of two people to complete an average of one site per hour, including travelling time. Each survey team could complete about 6-8 sites per day, or 40 sites per 5-day week. Surveys for a particular catchment or groups of catchments were organized by allocating enough teams to survey all the planned sites within a 2-week survey period (10-working days). The six pilot studies in Brisbane involved five survey teams. A total of 474 sites (257km of waterway in 521 sections) were surveyed over a two-weeks, with one team working an extra week. A medium size urban sub-catchment in Brisbane of 100-150 sites would require two teams for 12 working days, including training.
|
The software system consists of a set of 14 linked databases (one for each datasheet and one for the sections). It also includes programs for generating the ratings and for producing a comprehensive set of reports, maps, data downloads and other outputs. |
There is a linked database for each of the components and datasheets and a section database for storing the derived variables and summary parameters. There is a set of display forms that provide user-friendly access to the databases for entering, editing and verifying the data and for conducting various analyses and producing a wide range of reports. A comprehensive software system includes programs to generate the various derived ratings and other summary parameters. Once calculated the derived ratings are all stored in the databases for rapid access and use. The survey and derived ratings can be transferred to other databases, GIS or to various management systems that can interface with database information.
A wide range of outputs can be produced:
Audits of Stream Corridor Condition
These audits are produced as tabulated or graphic summaries for each of the attributes assessed in terms of condition. These summaries show the length of stream within the catchments or sub-catchments classified into a set of condition categories. These categories range from 'Very Poor 'condition (rating 0-20%), to 'Very Good' condition (rating 80-100%). The user can choose various other category systems or define their own. This provides an audit of the severity and size of the problem and quantifies the effort and resources needed for rehabilitating the streams.
Report Cards
Report cards summarize all the classification parameters for each section or for groups of sections (such as sections within planning areas or sub-catchments). The summaries include condition ratings, urban stream types, the type of channel habitat types present (pools, riffles, runs, etc.). Statistical summaries can be produced for channel dimensions and sediments for each type of channel habitat present. Key attributes for channel morphology, riparian vegetation and aquatic habitat are ranked according to their occurrence and relative abundance in the group of sites selected.
Outputs for GIS
The system is designed to download all the ratings to a GIS for each of the homogenous sections defined for the surveys.
Skeleton Maps
The package includes a mapping package that allows the user to generate simple 'skeleton' or sketch maps of the catchment surveyed. These maps provide cheap and fast preliminary maps that avoid the time delays and expense of GIS, which may not be available to community groups. This internal mapping system is the heart of the Decision Support System built into the package. The user can display all the classification parameters using coloured coded categories on these maps. The raw data collected during the surveys can also be displayed. A query system allows various combinations of data to be instantly displayed. For example, if the user is interested in identifying where fish populations could be reasonably established, a formula can be built which will select and categorize the suitability of sections which meet the criteria. The user may want to select sections which have pools deeper than 1m, have riparian zones which are natural and in reasonable condition, that are suitable for natural design techniques. The presence of riffles with gravel or cobble sediments and aquatic vegetation may also be important as spawning and rearing sites. Sections meeting these criteria can be displayed using the skeleton maps and rated according to their relative suitability. Most of the original survey attributes can be also be accessed using the databases and can be displayed in categories using the skeleton maps.
Comprehensive Statistical Reports
Comprehensive data summaries can be produced which summarizes all the data collected during the surveys for various groups of sites (sub-catchments, planning units, whole catchments).
Bed, Bank and Buffer Zone Modification Index
A classification system for urban streams was developed using the estimated naturalness and extent of modification to the bed and banks of the waterways, and three buffer zones – shoreline, middle and upper zones. The shoreline zone extended to about 10m, or two bankfull widths from the bank top. The middle zone extended for 20-30m from the shoreline zone to the 100-year flood boundary. The upper zone is the setback from the flood boundary to buildings or other structures. It is the permanent modifications to these zones that set limits and define the capability for rehabilitation or restoration of urban streams and corridors. Likewise the remnant zones may be too narrow to function effectively even if in a natural state or when fully restored. The adequacy of the width of the zones also limits the capacity for restoration or rehabilitation. Streams that are degraded can therefore only be improved to limits set by modifications or the inadequacy of the zone widths. This in turn can be used to set realistic and appropriate targets on stream sections of different type. A quick and easy way of addressing these issues was required.
A classification system was developed which includes the naturalness of each of the five zones and the adequacy of the widths of the buffer areas. Each zone was assigned a rating of 1-6. The waterways within a catchment were divided into sections that were homogeneous in terms of these urban stream ratings and condition of the various components. The modification index then defines the capability of the stream for rehabilitation or restoration. The condition ratings define the extent to which the full value and function of the various components are realised within the limitations imposed by the modifications. For example, a stream section with an urban rating of '66643' has a virtually natural bed, banks and shoreline and buffer zones but unnatural or narrow middle and upper zones. Another section with an urban rating of '34356' has modified bed, banks and shoreline zones but with natural floodplains and setbacks. If the riparian vegetation and instream habitat in both of these sections was in poor condition, there are major differences in the capability of the two sections for rehabilitation. The first section has a far greater potential for improvement. The second section had highly modified bed, banks and shoreline zone and this will limit the potential benefits of the rehabilitation. The full potential benefits will only be achieved if the modifications are removed, or natural channel design concepts are employed to reverse the channelisation.
Setting Baselines, Bench-Marking, Determining Trends, Rates of Change and Monitoring
The surveys and software systems have been designed to be repeated at regular intervals. The site locations and the reach boundaries are carefully defined so that the surveys can be repeated, either for the whole catchment or for some part of it. These surveys can be used to establish rates of change and trends and also to monitor improvements in condition following rehabilitation or other management initiatives.
|
The Decision Support System provides a comprehensive and powerful tool for selecting, rating and mapping the condition and suitability of various waterway sections for various purposes. The user can select from a wide range of in-built classification parameters. User-defined formulae can be used to derive various combinations of these parameters. The ratings produced can be displayed on maps using colour-coded categories to rate of the suitability, degradation or other criteria. |
Problems or issues can be identified and quantified in terms of size, severity and relative importance and priority established. The database, software system and various reports provide a comprehensive decision support system that is user-friendly and user-defined. The outputs and survey system are designed to provide quantitative audits of waterway and corridor condition in terms of absolute and relative (percentage) length of stream that fall within various condition categories. Having a simple way of identifying the problems, assessing their severity, and quantifying their size and importance, is the crucial first step in addressing the issues, setting priorities, and assessing the resources and strategies required to fix the problems. Quantifying the problem is also crucial when applying for funding. Likewise, having the ability to develop priorities based on comparisons between catchments and sub-catchments ensures that the funds available are used in the most cost effective and efficient manner. The comprehensive data included in the system provides these outcomes. The system was designed to be used not only by government employees, but also by community groups, schools and members of the public. The 'Skeleton Map' system was developed to cater for the many groups who may not have access to a GIS. The outputs, such as the 'stack diagrams' were designed to provide simple and easily understood summaries for a wide range of users.
The location and distribution of various attributes can be mapped, links established and relative rating or ranking for the selection criteria displayed in colour-coded categories. The 'Skeleton Map' system and GIS output have been designed to show the location and distribution of stream sections that meet a set of user-defined criteria. In simply terms this is designed to show: 'Where are the bad bits' and 'Where are the good bits'. This allows the most severely degraded sections, in terms of various problems, to be readily located on maps. Likewise it is important to recognize where the remnant areas in good condition are located so that they can be protected or used as reference area to define targets for rehabilitation of restoration. It is also important to be able to rate or rank the various sections and to display this on the maps. For example it may be better to focus rehabilitation efforts on the less severely degraded sections than on the worse cases which may be lost causes. The software system allows for this by mapping sections in various categories. It also allows the user to develop their own rating formulae using a combination of parameters to better define their requirements. Once again the output is generated in a eight categories for the ratings produced. Another important feature is that once a set of sections has been identified for a particular purpose, the user can step through the database system, and down to the original data collected at the representative sites. This provides a flexible and versatile system that provides decision support but does not dictate the decision outcome. It is a preliminary selection tool that meets the desired outcome for classification systems in simplifying complex issues, in providing a focus for inventory and action, and in examining linkages, interrelationships and dependencies.
|
Major Evaluation Trials have been completed in Brisbane for the survey techniques, software system, mapping procedures and the various outputs and reports. Brisbane City Council and community groups participated in these trials. |
The small pilot study originally planned for the study was expanded into a major evaluation exercise involving complete surveys of six catchments in Brisbane. The surveys were successfully completed within three weeks by five teams of two persons, which included staff from Brisbane City Council, University students and community group volunteers. A two-day training exercise was conducted before the surveys. Brisbane City Council and community groups trialed the software and DSS packages, and output system, and helped to refine these packages. The trial demonstrated the success of the package and enabled the time and personnel required for future surveys to be properly assessed.
Classification generally refers to the process of ordering of objects into sets based on their similarities or other relationships (Platts 1980). An efficient classification provides a simplification both in terms of reducing the range of groups that need to be considered and the set of attributes that are considered important in defining the groups. The desire for a classification system is driven by a need to simplify what are complex relationships between streams and their watersheds (USDA 1998). These relationships become much more complicated for urban streams and the major changes in runoff associated with urbanization. Classification provides a focus for action and facilitates effective management by highlighting the cause of changes in condition or stream type and the appropriate responses for rehabilitation or restoration. Management strategies may too often try to make all rivers perform all functions, or to try to force all rivers to fit a common morphological dimension or form (Rosgen 1996). Trying to recreate a historic meandering pattern or riffle/pool pattern may me ill-directed if the changes in runoff dynamics associated with increases in the imperviousness of the catchments is not properly considered. Urbanization changes streams in a fundamental way. Peak discharge and total discharge may increase two or three time through urbanization. Various types of streams are subject to different processes and are at different stages in terms of the evolutionary processes that formed and maintain them. Different types of streams in various parts of the catchment need to be managed with different priorities and approaches because they have different responses to management initiatives (Rosgen 1994).
The most effective classification systems are based on objective and quantifiable criteria that ensure consistency and reliability (Rosgen 1996).
Rosgen (1996) and USDA (1998) provide recent reviews of stream morphology classification systems. The key milestones are generally recognised as: -
An index of stream condition was developed in Victoria (Ladson et al. 1999). It uses assessments of hydrology, physical form, riparian zone, water quality and aquatic life individually and added together to produce and overall rating. Repeat surveys were proposed every five years for stream reaches tens of kilometers in length. It measures change from natural or desirable condition.
A geomorphological classification has been developed to provide a catchment-based scalar framework for examining human impacts on rivers and using geomorphology as a basis to assess river health and ecological applications (Brierley, Ferguson and Batten 1998; Brooks and Brierley, 1997; Brooks & Brierley, in press; Fryirs & Brierley, 1998). This method produces a classification into 'river styles' using valley width, channel planform, flood plain character, sediment storage and other attributes.
Anderson (1993 a, b, c) developed the Riverine Habitat Audit Procedures that were the precursor of the current system (discussed in section (1.6.3).
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:
Most previous studies have focused on morphological forms and processes for channels, stream corridors, drainage networks, flood plains and valley flats and sediment transport processes. In developing a decision support system for managing urban streams in terms of biologically important physical attributes the following classification targets were considered most relevant:
There is a need to classify reaches in terms of physical and ecological habitat types and their potential to support various types of plant and animal communities. The classification needs to address the current value and condition of the habitat and its potential after rehabilitation and restoration. This involves the assessment of:
This classification also allows the relative value of each reach section, tributary or sub-catchment to be assessed in terms of their actual or potential capacity to meet a defined set of habitat criteria for particular communities. The decision support system should enable a user to enter a set of key attributes and then to generate a report and map showing the location of suitable river sections and total length of stream within each catchment or sub-catchment that meets these criteria. Ideally the output should allow the identified reaches to be placed into suitability categories in terms of the extent to which the criteria are met, or to define various constraint categories that show the rehabilitation options and priorities.
As well as identifying the type of habitat present there is a need to identify its current state or condition, as this qualifies its value and utility. There is a need for an audit of environmental condition and values. Ideally this condition should be assessed using reference sites to allow a relative assessment of the extent of degradation or loss of utility, value or function (Fig. 1). Such a condition assessment is an integral part of the assessment of asset, audit and value assessment. It also may be used to define the target or scope for rehabilitation or restoration. However it may not be possible to return the stream to its original condition and therefore to regenerate all of its original utility or functions (Fig. 2). In some circumstances this may be counter-productive. There is a need to identify the capacity or capability of the stream section as a final threshold after rehabilitation or restoration (Fig. 3). This includes recognising a limit or threshold to the restoration that can be accomplished. Also the target type of channel or channel dimensions may be different from that which occurred historically.
Degradation can be regarded as a short to long term process often being dependent on major episodes of change triggered by some change in the catchment affecting hydrology or sediment supply and transport processes (Fig.1). The pathways of change in time may be different in each stream section, and there may be natural recovery processes. The process of degradation may be assessed as a percentage loss or value or utility from some original pristine and fully functioning state to the point where virtually all of the values and functions have been lost. 'Snap-shot' surveys can provide an estimate of the position of a particular reach in terms of the extent of degradation as a percentage change from some original condition or a s a percentage loss of value or utility. However, these 'Snap-shot' surveys can also provide an indication of the processes involved and the trend and rates of change if they are repeated regularly.
Rehabilitation and recovery activities may reverse the trend and re-establish part or all of the value or utility lost through degradation. In this context rehabilitation is the re-building of riffles or other habitat structures in the stream or activities to increase the habitat complexity within the stream without completely rebuilding it. Restoration refers to the major works which may often involve efforts to address the cause of the original degradation (Goodwin, Hawkins & Kershner 1997), including recreating the historic hydrologic characteristics and making predictions are made of expected responses by the geomorphic and vegetative components (Toth et al. 1995). Rehabilitation is the 'patch job'; restoration is the 'rebuilding or reconstruction'. In this sense, restoration may involve a shift to a different basic type of stream as well as improvements in value and utility, whereas rehabilitation normally does not involve a major shift in type. The term restoration may include rehabilitation initiatives. The following definitions (USDA 1998) provide a good understanding of the differences:
"Restoration is the process of repairing damage to the diversity and dynamics of ecosystems. Ecological restoration is the process of returning an ecosystem as closely as possible to pre-disturbance conditions and functions. Implicit in this definition is that ecosystems are naturally dynamic. It is therefore not possible to recreate a system exactly. The restoration process reestablishes the general structure, function, and dynamic, but self-sustaining behavior of the ecosystem."
"Rehabilitation is making the land useful again after a disturbance. It involves the recovery of ecosystem functions and processes in a degraded habitat. Rehabilitation does not necessarily reestablish the pre-disturbance condition, but does involve establishing geologically and hydrologically stable landscapes that support the natural ecosystem mosaic."
|
|
|
Figure 1. Time series changes in relative value and utility of streams, channels, riparian zones and floodplain/valley flats with degradation, rehabilitation and restoration. |
|
|
Figure 2. Degradation pathways may involve a transition to different types of streams. Likewise rehabilitation or restoration may not involve retracing the degradation pathways but may lead to new endpoints in terms of stream type. |
|
|
Figure 3. The effects of constraints in limiting the capacity and capability of various stream types for restoration. |
"Reclamation is a series of activities intended to change the biophysical capacity of an ecosystem. The resulting ecosystem is different from the ecosystem existing prior to recovery. The term has implied the process of adapting wild or natural resources to serve a utilitarian human purpose such as the conversion of riparian or wetland ecosystems to agricultural, industrial, or urban uses."
"Restoration differs from rehabilitation and reclamation in that restoration is a holistic process not achieved through the isolated manipulation of individual elements. While restoration aims to return an ecosystem to a former natural condition, rehabilitation and reclamation imply putting a landscape to a new or altered use to serve a particular human purpose."
Regular monitoring and management initiatives specific for each stream types need to be defined.
There is a need to identify key constraints limiting the scope of restoration or rehabilitation of various stream types and the long-term outcomes. Major constraints such as imperviousness may severely restrict what can be achieved irrespective of what is done in the corridor. Schueler (1995) highlights the importance of imperviousness in modifying runoff dynamics and stream power in urban catchments. The increased imperviousness of urban catchments when above a threshold of about 10% affects stream morphology and profile, water quality, water temperature and ultimately stream biodiversity (Schueler 1995). The thresholds for each of these impacts provide a reasonable foundation for classifying the potential stream quality in a watershed based on the amount of impervious cover. He proposed that streams with 1-10% impervious cover are classified as 'Stressed'; 11-25% impervious cover as 'Impacted' and 26-100% impervious cover as 'Degraded'. Imperviousness ultimately restricts the potential stream quality that can be achieved, even if restoration of the stream corridor is undertaken. These problems may be addressed through source controls, better urban planning better site design (Schueler 1994). Various types of stream may affected in different ways and classification is important. Such constraints ultimately limit the capacity to restore urban stream corridors.
Permanent changes to the hydrology, sediment supply and transport processes ultimately set limits to restoration. Channelisation, alterations to the channel profile and concrete linings also reduce what can be potentially achieved. Urban streams can be classified in terms of the type and extent of these modifications on the bed, bank, floodplain and setback to buildings and other structures in the transitional upland areas. Goodwin, Hawkins & Kershner (1997) suggested two zones: one corresponding with the active floodplain that is frequently inundated, and the other extending from the active floodplain to the valley wall. Schueler (1995) proposed three lateral zones – streamside, middle and outer zones as part of a buffer system for urban streams. Each zone performs a different function, and has different effective target width, vegetative targets and management scheme in terms of allowable uses. Ultimately it is the modifications to the width of these buffer zones and their vegetative cover and land use that set restrictions on what can be achieved in the adjacent channel and in channel areas downstream. Direct modification to the bed of the stream or to the banks in terms of complete replacement with trapezoid concrete linings or other armouring through rip-rap and erosion control devices on the bed and banks also restrict the potential or capacity of the stream functions to be fully restored. Additional capacity for improvement can only be achieved if these major constraints are removed by restoring the original bed and banks and the original hydrological characteristics. The application of 'soft-engineering' approaches to stream corridor restoration aims to shift the threshold of maximum potential for restoration by removing the constraints inherent in channelisation and the restoration of natural pool and riffle sequences (Newbury and Gaboury 1993). The urban stream classification that has been developed for the 'Waterway Classification' datasheet described later in this report is based on this concept of assessing the condition and minimum width, naturalness and artificial modifications to the bed banks and three streamside buffer zones.
Water quality and point sources of pollution within the catchment also act as constraints and limit the potential or capability of the stream to be restored or rehabilitated. These restrictions may apply well downstream of the source of the water quality problems. This emphasizes the need to consider the capacity of the stream corridor for restoration within the context of the entire catchment as expressed through the drainage network in developing management and restoration priorities.
The key to understanding how various constraints may affect waterways and corridors is to develop links for the type of interaction and the processes involved:
Effective planning and management also focuses on defining opportunities and knowing when they will occur and how to influence their outcomes. All artificial channel structures have a limited and defined life, and there may be regular maintenance requirements and schedules. Focusing on these opportunities and having plans and budgets prepared is a good strategy for introducing 'soft engineering' approaches and rehabilitating urban streams. Good local manuals are often available (e.g. Brisbane City Council 1995).
There were two general approaches that could have been adopted in developing a classification system for management of urban streams:
The second option was adopted for this study. The major reason was that there was a need to use the component classifications as well as an integrated classification and to use the raw data as well as the derived data as part of the decision support system. Gurnell, Angold, & Gregory (1994) analyzed 140 international publications on river corridor classification and found that robust operational classification schemes were based on a wide range of information types with a hierarchical structure incorporating different types and resolutions of information suitable to support classifications for different applications.
Gurnell, Angold, & Gregory (1994) recommended that data handling should maintain separation between raw data and their derivatives. This principle was adopted for this study. The data were compiled into databases and the derived ratings and summary variables were added after analysis. All data are then available to the user in a hierarchical and linked system. The derivation of the ratings and summary parameters are always made explicit.
The 'State of the Rivers' (SOR) methodology was developed by the author of this report (Anderson 1993 a, b). These methods have been applied as the 'Anderson' method or the 'Riverine Habitat Audit Procedures' (RHAP) to 16 catchments in Queensland (3400 sites) and to 10 catchments in NSW (2000 sites) (Table 1). A number of reports have been completed for these methods (Anderson 1993c; Phillips and Moller 1995; Telfer 1995; Carter 1997). This system was modified and upgraded to meet the objectives of the project for urban streams.
The SOR and RHAP methodologies have been used for the 'State of the Rivers' program in Queensland (Anderson 1999b), for 'State of the Environment' reporting, for developing integrated catchment management plans in Brisbane. RHAP has also been used for environmental impact assessment studies (Anderson 1998a) and for environmental flow determinations (Anderson 1998b). The derivation of the methods and their validation through trials in the Maroochy River has been documented (Anderson 1993a, b). The Department of Natural Resources in Queensland is currently re-surveying the Maroochy catchment to monitor changes in the river corridor and to assess the outcome of various rehabilitation and control activities undertaken in the catchment (personal communication, Glen Moller, DNR). This organisation is also currently investigating the use of the 'State of the Rivers' data in Queensland for developing an integrated waterway classification system for all rivers in Queensland. This classification will be used to assist in the development of conservation priorities and sustainability of future water resource developments. It was appropriate to extend these techniques into urban catchments to benefit from the consistent approach and universal ratings. Many urban catchments have sections that are in good condition with essentially natural bed, banks and flood plains. The adoption of a similar approach allows continuity, especially where cities are sited on the floodplain of river systems with rural land uses in the headwaters. This is particularly true in Australia (e.g. Brisbane, Perth, Melbourne and Sydney).
These methods employ many of the techniques developed overseas for assessment of riverine habitat type and condition (USFWS 1980 a, b; 1981), for riverine corridor restoration and for river health assessment. These methods are generally similar to the recently developed visual assessment protocol for streams (USDA-NRCS 1999). However these methods use a direct assessment scoring system (scores from 1 to 10) for various components, without collecting raw data. The protocol also includes a mixture of habitat, hydrological and biological data. However it is less comprehensive than the system described in this report and does not specifically address urban stream classification and management issues.
| Table 1. Catchments in Queensland and NSW where the 'State of the Rivers'/Riverine Habitat Audit Procedures have been completed or are in the final planning stages. |
| Queensland Catchments | New South Wales Catchments |
| Maroochy | Tweed |
| Upper Condamine | Brunswick |
| Herbert | Richmond |
| Dawson | Clarence |
| Mary | Lachlan |
| Lockyer | Macquarie |
| Bremer | Port Hacking |
| Logan | Bega |
| Tully/Murray | Lake Macquarie |
| Cooper Creek | Border Rivers |
| Border Rivers (several rivers) | |
| Caboolture | |
| Mooloolah | |
| Burnet | |
| Mitchell | |
| Gilbert/ Norman | |
| (8 further catchments in preliminary planning stage) |
A workshop, hosted by the Brisbane City Council, was held in Brisbane in November 1996. It was attended by 65 participants from Local and State Government representatives, various interest groups and community groups from around Australia. Keynote addresses were given by Ms Rachael Barker from the Christchurch City Council, and Dr. Bruce Hamilton from NIWA in New Zealand. Geoff Hunter from the Blacktown City Council presented a paper summarizing the findings from a trip to USA funded by a Churchill Fellowship.
Draft survey datasheets were distributed to the workshop participants for comment. The workshop focused on the following topics:
A report on the outcomes of the workshop was circulated to the participants and the recommendation was incorporated into the project.
A newsletter ('Lets Fix Urban Streams') was circulated to all Local Government Organizations in Australia and to various other organizations to promote the project and to request feedback on the major elements of the project.
The project was developed with the cooperation and assistance of the Brisbane City Council. This was important for ensuring that a practical system was developed that would meet planning and management requirements.
A consultation tour of Local Government Agencies in NSW, Queensland and ACT was undertaken to finalize the design of the system and to clarify how the system would be used in managing urban streams.
Pilot studies of six catchments were undertaken in Brisbane in 1996-1997 to develop and refine the methodology. The outcome from these studies is reported separately (Anderson 1999b).
The major recommendations from the workshop were:
The tour of NSW, QLD and ACT local government groups confirmed that specific purpose surveys were required. Existing data were not consistently available throughout entire catchment areas. The proposed sampling strategy using 'homogeneous stream sections' could not be applied to most areas because planning units and other areas contained a variety of stream types. Likewise the planning units did not necessarily follow catchment boundaries, nor were the boundaries located at the nodes in the drainage network.
The sampling and survey design are similar to that described for the SOR procedures (Anderson 1993a,b). The survey design was based on the concept of a 'homogeneous stream section' that is a reach which has similar physical features in the channel in terms of stream type, channel modifications, channel habitat types (pools/riffles etc.) and instream and riparian habitat types and condition. The process is to continually subdivide the streams within a catchment until this criterion of 'homogeneity' is satisfied at the level of detail required for the survey. This provides a stratification process and also means that information collected at one or more sites, which are representative of the stream corridor within the defined section, can be applied to the entire section. The process of defining the sections is shown in Fig. 4a. A catchment map for the pilot study in Norman Creek, Brisbane, is shown in Figure 4b.
The surveys are focused on major streams, that is streams that have a well defined channel and that are shown on 1:10,000, 1:25,000 or 1:50,000 scale topographic maps. There may be a need to exclude many of the minor tributary branches as temporary channels, particularly if the channels are ill defined. The scope and resolution required for each study have to be defined to keep the surveys realistic in relation to funds and resources available. The data collected and ratings produced only apply to the stream length surveyed (major streams). The minor streams are designated as 'not-surveyed'.
The definition of the boundaries for the homogeneous stream sections is first carried out as a mapping exercise. This is usually done as a workshop involving local government personnel and representatives from local community and other interest groups and local residents. Topographic maps, planning maps, catchment management maps, as well as any other maps (including local street directories) can be used to define major boundaries in the waterways and stream corridors.
Section boundaries are first established at the major junctions within the drainage network, separating the tributary section and the main channel sections immediately upstream and down stream from the junction (Fig. 4a).
Planning unit and other administrative boundaries are used to ensure that the ratings produced relate to the way the waterways are currently managed.
Boundaries are established at major artificial dams and weirs; at natural barriers such as waterfalls, rapids, instream wetlands; and at other locations of barriers and obstructions. Dams and weirs act as sediment traps and lead to changes in flow regimes downstream. The entire reservoir, weir pool or natural lake is designated as one or more distinct sections because of the different habitat upstream of such barriers. Dams, weirs and other barriers such as waterfalls and rapids impede fish movement and the passage of other animals often leading to changes in community structure.
Altitude itself is an important parameter affecting local climate and the distribution of many plant and animal species. Catchment slope affects run-off dynamics and this influences channel morphology. Stream gradient is also of fundamental importance for stream classification as it affects flows, sediments and many other physical and biological features of streams. The altitude categories chosen will depend on the catchment - its topography and location. Likewise the categories for stream gradients will depend on local conditions. Some of the obvious boundaries are the upstream and downstream ends of gorges where streams descend rapidly from coastal escarpments, waterfalls and rapids, tidal limits and the upstream limits of the broad alluvial floodplains. Climatic and rainfall maps may also help to identify suitable boundaries.
|
|
Figure 4a. Sub-sectioning of stream into homogeneous stream sections using the drainage network nodes and altitude. Two methods of displaying the sections using GIS are shown. The 'Spaghetti' method displays widened corridors and the 'Pizza' method uses the local sub-catchment areas to show the stream sections (enlarged image) |
|
|
|
The geology and soils of the catchment and the geology beneath the streambed are important in affecting the channel morphology and other features of streams.
Sub-section boundaries should be established at major discontinuities in existing vegetation, for example conservation areas (National Parks, State Parks and conservation areas and other reserves), forests, plantation areas, and boundaries of natural vegetation types. Major land use types, and urban land use and planning boundaries and waterway planning maps may also be used to identify different stream sections.
Land system and elevation boundaries can also be used to sub-section the streams. Again the locations where the boundaries between different land system types intercept drainage lines can be used as the location of stream section boundaries.
Boundaries are also established at the tidal limits and between permanent and intermittent stream sections.
The location of discharge points and major disturbance sites are also important. Examples of such sites and disturbances are: -
These represent potential sources of disturbance to conditions downstream.
These disturbances including channelisation and bank protection works.
Water quality and biological monitoring data may also be used to define further boundaries.
Many other attributes may be used to allocate further sub-section boundaries. It is not realistic to define a fixed procedure and list of attributes that are used because the information available varies between catchments and may be inconsistent over the entire catchment. The types of information likely to be relevant are: tenure, gravel or sand extraction sites, weed infestations, tree planting, and the various zoning and urban development plans for the area.
Obviously this process could continue indefinitely and limits need to be set in terms of the scope, resolution required and resources available for the study. At some stage there has to be a decision made about the scale that is to be applied and ultimately how many sites can be reasonably surveyed with the resources and time available. Once again this is a pragmatic decision. The following points are relevant:
Following the mapping exercise, reconnaissance surveys are conducted throughout the catchment to validate and further refine the definition of the homogeneous stream sections. One or more teams of two people normally conduct these surveys. Sites representative of the type and condition of the stream, banks and floodplain/valley flat areas are also chosen during these surveys. The sites are delineated as reaches of variable length, usually including two meander wavelength lengths. Sites are also chosen to include appropriate representatives of the range of channel habitat types present in the section (pools, riffles, runs, etc.). Where possible the designated reach should include the deepest pool in the area and at least one riffle/run/glide type habitat. This ensures that the full range of sediment types, depths and habitat features are included for the section surveyed. Reaches are usually about 50m long but may be 200m long in larger streams. The other criterion is that the entire length of stream in the reach must be visible from the one vantage point in the reach. Many of the attributes that are assessed during the surveys refer to the percentage of the length of the banks or bed in the reach that is covered with aquatic vegetation for example, or forms a bar, or has canopy cover present.
The final and most limiting criterion in selection of the sites is access. The reach must be reasonably accessible and easily surveyed. This means choosing sites that are within 50-100m of the vehicular access.
Sites are normally located well upstream of bridges, culverts and other obstructions that may affect the channel morphology. Further details about the selection of the sites (reaches) are provided in Anderson (1993a).
Usually the first data sheet (Site Description) is completed during the reconnaissance surveys to define the location of the site using GPS readings and map coordinates. Sketch maps and site descriptions using local landmarks are used to define site locations and reach boundaries. A standard set of photographs is also taken at this time to reduce the time spent at each site for the main survey and to the number of cameras required. Details about this datasheet are provided later in this report.
Each reconnaissance team of two people can cover about 20-30 sites in a day, depending on access and time taken between sites.
Once the sub-sections have been delineated, the local sub-catchment element draining into each section is marked on the topographic maps and becomes part of the GIS mapping that is normally conducted for the surveys. This enables links with the catchment to be established in terms of land use, imperviousness, etc. The entire catchment area is included in this process Two approach have been used in the past for displaying the ratings for each sub-section that is defined. These methods have been colloquially referred to as the 'Pizza' and 'Spaghetti' methods depending on whether the sub-catchment element linked with each stream section is coloured to display the rating categories ('Pizza') or just the expanded stream corridor ('Spaghetti') (Fig.4a).
It was difficult to directly incorporate existing information and data sets into the DSS. There are sampling issues, inconsistencies and variability in scope and coverage. The database system was developed to record this existing information so that it can be viewed within the same sampling framework as the present study. Most of the summary data attributes can also be plotted and downloaded. This encourages the development of links with existing data but avoids the problems of inconsistent and incomplete data when deriving the ratings. Hydrologic data, channel modification information, natural channel design parameters, information about imperviousness, and other features of the catchment can be recorded. These parameters are not directly used in calculating the ratings but they can be mapped. Once the 'homogeneous stream sections' have been identified it can be used to provide a sampling framework to link with existing data. Sampling sites for water quality and biological monitoring can be associated with particular stream reaches and the various data sets and summary parameters produced by the DSS. For example fish distribution, abundance or biodiversity could be linked with bed sediments, channel habitat types and dimensions and various condition ratings such as riparian and aquatic vegetation, aquatic habitat or conservation value ratings. Likewise it may be possible to apply ratings derived from water quality data, or from biological assessment of river health to each of the 'homogeneous stream sections'. This would provide a direct link with external data sets and information.
The major criteria for developing the survey methods and survey design were similar to that required for the RHAP system (Anderson 1993a,b). The workshop confirmed that a similar approach was required. The major design criteria were:
The pilot surveys in Brisbane were conducted by teams that included University students, community group members and staff of the Brisbane City Council (Anderson 1999b).
The following list of equipment is required for each survey team (2 people):
A training program is conducted over 2 days (Anderson 1993 a, b). The datasheets and survey methods are covered in the first morning. The afternoon of the first day is spent with all trainees jointly surveying the same local sites under supervision. On the next day the teams are send out to start collecting data. An experienced person spends at least half a day in the field with each team. The datasheets are checked and any problems addressed. Additional support and further checking occur during the first week of the surveys.
There are 13 survey components, each with its own datasheet. The datasheets are shown in Appendix 1. Details of the datasheet design and the development of the attributes for the first 9 datasheets are provided in Anderson (1993 a, b). There have been only minor changes to these original datasheets that were used for the rural surveys, and none of these were made specifically for surveys of urban areas. The desire to change the datasheets for urban streams was counteracted by the benefit of having a consistent assessment method, as many parts of urban catchments may be quite similar to rural areas. This enabled rural and urban catchments to be realistically compared.
The datasheets were developed from those used by Anderson & Morison (1979) and Mitchell (1990). The following aspects were important in designing the datasheets:
No coding sheets are used
The datasheets were designed to eliminate the need for coding sheets. Inexperienced personnel find it difficult to use coding sheets, particularly when the category codes vary between attributes. It is far easier for the survey person to enter actual percentage estimates or measured values than to have to look up a coding sheet and to enter a code for the category for the estimate or measurement. The raw data estimates are stored in the databases and the analysis programs assign the category as required for estimating the ratings. The only problem within this approach is that it may imply that the accuracy of the estimates is beyond what can reasonably be expected. However, cover estimates are only required to the nearest 5% and there is broad categorization in the analysis. The elimination of coding sheets means that all of the possible values or attributes have to be shown on the datasheet itself, and this makes the layout somewhat 'cluttered'. However the users soon become familiar with the layout.
All the datasheets, except for the cross-sections can be completed by one individual. At each site the datasheets are divided between them according to their individual preferences. The team members must however work as a team in completing the cross-section datasheet.
Graphics are used extensively to simplify the data collection using the datasheets.
There are many data types on the datasheets, from actual measurements of width and depth to estimates of distances and cover areas. Various rankings and category allocations are also used. Instructions are provided on the sheets to identify the requirements for each attribute. Sequence numbers (in boxes) are provided as a guide to the sequence in which each data sheet should be completed. Further details about the data types are provided in Anderson (1993 a, b). The datasheets are a major topic of the training program.
Subjective and Objective Ratings
Most data sheets include both subjective ratings that are made by the survey personnel using their own judgement. Objective ratings are generated after the survey has been completed in the analysis package. The objective ratings are normally used in the outputs, but the subjective ratings are also available as raw data. The subjective ratings are mostly used for checking purposes to verify unexpected or unreasonable objective ratings.
A unique code is required for each site surveyed. The numbers allocated to each site may be done purposely in terms of the drainage network or planning units, but this is not necessary. The numbers allocated are simply identification codes. The same applies to the subsection numbers. The basin number is that assigned to each basin by the Australian Water Resources Council. For the purposes of the databases each site in a basin is allocated a 'trip-code'. This is generated at the time of data entry in the following way:
The survey number is used to allow the surveys to be repeated for monitoring and to follow the progress of management and rehabilitation initiatives. This survey number is added during data entry.
Bank top, Banks - From a measurement point of view the banks extend from the edge of the water surface (or the 'water mark' if the bed is dry) to the bank top, that is to the point of inflexion of the channel profile. The bank extends to the edge of the active hydrologic floodplain or valley flat. In some areas there may be complex terrace structures. These terraces are usually included, except when it is obvious that that the more proximal terraces are old and not actively worked by the stream.
Water Surface – The water surface was used for taking depth measurements for the cross-section measurements and for assessing the instream organic debris cover available for fish and invertebrates. When the bed is dry assessment is made at the 'water mark', i.e. assuming that the channel was filled with water to the 'water mark'.
'Water Mark' - The concept of a 'water mark' is used to provide a reference point for standardizing the channel measurements and for marking the boundary between the lower and upper banks. A mark is left on the bank at the 'normal' inundation level for the stream. It is delineated as either at the edge of the terrestial grasses or other vegetation, which cannot tolerate frequent inundation, or an obvious point of erosion or substrate differences along the bank. Anderson (1989) describes its origin, definition and use in more detail. The concept was originally developed by Woodyer (1968)
Riparian Zone - The riparian zone is an interface between the stream and the surrounding land. The vegetation in it is different because of the influence of the stream in increasing available moisture, flooding and soil characteristics. The vegetation is important because it contributes organic debris, stabilizes the banks and provides shade, and cover for the instream communities.
It is difficult to derive a precise definition of a riparian zone, particularly for untrained staff. There are many possible definitions. The most relevant feature is the vegetation cover and diversity. Ideally it should be possible to identify such a zone irrespective of disturbance or condition, and it should have a fixed width for each size and type of stream, relating to how far and how often the stream or river floods or reaches bank-full stage. However, this is not possible in a 'snap-shot' survey of this type conducted by non-exerts.
The general perception of there being a strip of vegetation along the edge of the stream, which is different from the rest of the vegetation in the landscape, is generally obvious and recognizable by most people. The key attributes for assessment are the width of the remnant zone, its longitudinal extent and the condition of its vegetation communities in terms of density (= % cover) and invasion by exotic species. It is also clear that the width of this zone will vary markedly depending on location, the slope of the valley flat, soils and also the extent of disturbance. The options of setting a fixed width, or trying to define the width of the original undisturbed zone or exactly defining the width of the riparian zone in some other ways are impractical and would lead to errors. Untrained staff may not be able to apply these definitions.
The focus is on identifying the remnant riparian zone, which is more easily recognised because of its distinctive type of vegetation. The cover and diversity of structural vegetation types (trees, shrubs, herbs, grasses, etc.) is then estimated within the boundaries of this remnant zone. The riparian vegetation on the left and right banks are assessed independently and separately recorded. The aim is to identify how much of the original function and value has been retained by comparison with an expected width (usually 50m), diversity and cover. The ratings can be scaled using reference site in good condition.
The datasheets are shown in Appendix 1. A database was established for each datasheet.
This datasheet is used to define the location of the sites and the reach boundaries. The general location of the site is defined using map coordinates or GPS defined locations. The location of the site is also defined and a sketch map is used as a further guide to the location of the site. Sites can be full survey sites or sites where only the photographs are taken. The presence of stream gauges, water quality or biological sampling sites within the reach can also be recorded.
A standard set of photographs is taken at each site. This includes photos taken facing upstream and downstream, and facing the left bank and right banks (lateral views). Photographs are also taken of the entire reach and other relevant features at the site. The film and shot numbers are recorded in the database so that the slides can be labeled and archived. These photographs are an integral part of the surveys and they are invaluable for checking purposes. Photographs taken during previous surveys are now being incorporated into CD-ROMS that include other data summaries and other information.
This datasheet is designed to classify the site and section it represents in terms of the features on the floodplain and valley-flat, that is in the area adjacent to the riparian zone. It includes information on the water level at the time of sampling, channel pattern (map and local scales), estimated floodplain/valley flat widths, local land use, local disturbance, local vegetation type, floodplain features and land tenure. The local vegetation type refers to the original vegetation type that occurred in the area not to the current vegetation.
Subjective Rating
The subjective overall disturbance rating (scored 1-6) includes and assessment of the extent of modification of the valley-flat and shoreline vegetation and the extent of invasion by exotic species.
The aim of this datasheet is to classify the segments of the reach into the following broad types of channel habitat: -
These types represent the broad range of aquatic habitats present and the diversity of sediments, flows, depths and general habitats present in the reach. This classification is important for both the physical and environmental condition of the stream. The classification was derived from that developed by Anderson and Morison (1989). The diversity of habitat types present in the area is recorded as percentages of the section length. The average length, width, and depth of each type are also recorded. The total length of the reach selected for the survey is also recorded.
Cross-sections are taken to characterize the channel size and shape and the channel profile for the various channel habitat types present in the reach. Sediment samples are also taken at several points across the bed and also on the lower and upper banks. The particle-size composition of the sediments is determined by visual inspection using the estimated percentage contribution of each particle-size fraction to the total volume of inorganic materials present (Anderson and Morison 1989).
Channel shape, morphology and dimensions are fundamentally important for classifying the aquatic habitats in relation to physical and biological attributes. Stream invertebrate distribution and abundance is very much influenced by the type of substrate present and the relationship between flows depths and substrates. Bank and bed sediment particle size composition also dictates the response of the stream to various disturbances. The cross-sections also provide base-line information for follow-up surveys when changes in channel dimensions may be detected. They are also important for understanding the processes affecting the stability of the channel (e.g. headward erosion of the bed).
The techniques and data sheets have been adapted from Anderson and Morison (1989). The sediment size classes are those of the Standards Association of Australia, Australian Standard 1726 - 1981, SAA Site Investigation Code (gravel = 2 - 60mm; sand = 0.06 - 2mm; and silt/clay = < 0.06 mm).
Measurements are made from bank top to bank top at two or more cross-sections within each survey reach. Starting the measurements from the left bank, facing downstream, standardizes the measurements. Recording the locations where the depth changes significantly across the stream, maximizes the information collected. Extra sheets can be added but eight measurements are normally sufficient to characterize the profile. The width, height and slope at the base of the lower and upper banks are recorded on each side. Sediment particles size estimates are made for each depth measurement and for the lower and upper banks using the 'visual volumetric method' described on the datasheets. Sizes of well-known objects corresponding with the boundary sizes of the classification, are shown to help classify the sediment particle sizes.
The focus of the surveys is on the extremes within each reach - the point of maximum depth and minimum flow (pools), and the point of minimum depth (highest bed height across the stream) and maximum flow (riffle or run section). At least one cross-section should be taken through the maximum depth of a pool and shallowest point in a riffle/ run habitat. Additional cross-sections should be taken for the other channel habitat types present in the reach. Flows can also be recorded.
A portable echo sounder is used for taking depths at sites too deep to wade. The transducer is fitted into a 'kick-board' and the cable lengthened to about 25-50m. With one team member on each side of the stream a tape is attached to one side and a rope to the other. The transducer and float can then be pulled across the stream and a set of depth and distance measurements recorded from the left bank. The echo sounder can also be used to find the deepest point in the pool.
Some sites will have to be surveyed by boat or by swimming, again using an echo sounder. At deeper sites a grab or other type of sediment sampler is used to collect sediment samples.
When there is no water at the site at the time of the sampling depth measurements are made using a tape stretched across the channel at the 'water mark'.
Two data sheets are used to assess the condition of the bed, bar and banks. These data sheets were developed from those of Anderson and Morison (1989), those used for the study reported by Mitchell (1990). The ideas of Geoff Eades, Department of Primary Industries, Queensland were also incorporated (personal communication).
The time frames for channel metamorphosis and change in response to changes in sediment load and flow regimes are complex. Bankfull discharge events are seen as the primary flows for shaping the channels but larger catastrophic flood flows are also important as is their timing in relation to the recovery period at lower discharges. The dynamics of the processes are therefore difficult to assess using single 'snap-shot' surveys. The surveys focus on the dominant process at the time of the sampling - that is, whether the banks are eroding, or whether sediment is accumulating, forming bars and filling in the channel. What is observed will obviously depend on the time that has elapsed since the last major flood or bankfull discharge. The survey can be used as a baseline against which to assess future changes for follow-up surveys, which can provide information on the trends and rates of change.
Estimating the percentage of the length of bank (upper and lower) assesses bank condition in the reach (left and right), that is stable or unstable (eroding, aggrading, slumping) at the time of the survey. The recorded percentages must total 100%. The location of the instability (bends, obstacles etc.) and the local factors affecting stability are also assessed for each type of instability present to help to identify the processes involved. Slope and shape are also classified as ranks of the type present. The factors affecting stability and the artificial bank protection measures installed are also recorded as ranks.
Subjective Rating
Subjective ratings of the condition of the left and right banks are made in terms of the overall instability, and the susceptibility to erosion.
The type of bars present and the total area occupied by bars are assessed, using the 'water mark' as the reference point. The dominant particle sizes, angularity, shape, and surface deposits on the bar substrate are also recorded. Bed compaction is assessed. The factors affecting stability and the controls stabilizing the bed are recorded as ranks.
Subjective Rating
The overall stability of the bed is rated using the criteria present on the datasheet.
Passage and Barrier Classification
The suitability of the site for general fish passage is estimated for water levels prevailing at the time of the survey and also for water levels at the 'water mark'. The type, height and water stage required to over-top the barriers, and effectively remove the barrier effects are also recorded.
Riparian and aquatic vegetation is recognised as one of the most important features for assessing the condition of streams. Both riparian vegetation (Harris 1988) and aquatic vegetation (Holmes 1989) have been used for stream classification. Riparian vegetation acts to:
Riparian and aquatic vegetation is assessed in terms of percentage cover for various growth forms and the cover and presence of key local native and exotic species. The percentage cover by exotic species is also assessed. The methodology for the assessment was modified after Anderson and Morison (1989). Foliage structure and cover (foliage density) were used as the assessment methods because of the limited knowledge and experience of the survey staff. An optional checklist of locally important species is also included for both the left and right banks. This checklist is compiled during the planning stage and keys, photographs and other identification materials are provided to each survey team. A workshop session on identifying the plans is included in the training program.
The following definition is important for the surveys:
The person conducting the surveys defines the remnant zone and its width. If the zone cannot be clearly defined then the assessment should be made for a uniform strip 5m wide. The maximum possible width is 999 m. The riparian zone width will also vary with stream order and stream corridor width. If the riparian zone width has not been altered, and there are trees shrubs and grasses through the area then the width is normally set to 50m. This applies even in small first-order streams. A maximum width of 50m is used when deriving the rating, and this ensures that sites in pristine condition have this minimum width when deriving the ratings. Data is collected independently for the left and right zones. The estimated width of the original riparian zone, and a modification index (see Appendix 1) are also be entered and this information provides a way of checking the ratings.
The percentage of the remnant riparian zone covered by various structural vegetation types is recorded for the left and right banks, as well as the percentage of this cover occupied by exotic species. The vegetation types may overlap and so the total cover will normally exceed 100%. The total percentage cover of the entire remnant zone by weeds and exotic species is also recorded. The species in the checklist are recorded as rare (<50%) cover, or abundant (>50% cover). A checklist of common weed species is also included.
The aquatic vegetation is assessed as a percentage cover of the wetted surface area in the reach, that is the area covered by water, or the area below the 'water mark', if water levels are low. The total for the percentage of the bed devoid of vegetation, and area covered by submerged, floating and emergent vegetation, should total 100%. The individual species and vegetation types are recorded as percentages of the total cover for the general type (i.e. the contribution of each species as a percentage of the cover for the type of vegetation).
In each case provision is made for recording additional species from those shown on the sheets. The vegetation datasheet has to be modified for each survey to record these other species and the checklist species.
Some of the attributes collected on the other sheets are important for assessing aquatic habitat. These include:-
Inorganic debris covers of various types are also important for fish and invertebrates. Instream cover in the form of logs and branches provide shelter and attachment points, and also increases the diversity of flow and depth in the channel. Bank and vegetation cover also provides shade and shelter for the stream. The methods of assessment have been adapted from Anderson and Morison (1989) who developed their methods by reviewing overseas literature and the range of techniques used in Victoria. There is extensive literature on the relationship between fish populations and habitat parameters (e.g. Platts, 1979; Hubert 1988; Kruse et al. 1997; Hubert, 1997; Copp 1989; Platts and Nelson 1985; Heede and Rinne 1990; USFWS 1980a; USFWS 1980b and USFWS 1981). Relatively few such studies have been published in Australia (e.g. Davies 1988; Arthington 1992; Arthington et al. 1983). There is also an extensive literature on the habitat requirements of macroinvertebrates (e.g. Culp et al. 1983; Cushing et al 1983; Boulton and Lake 1990; Naiman et al. 1992) including attempts to predict invertebrate community types present in an area from stream measurements (Wright 1995; Wright et al. 1984).
The major objectives are to classify the aquatic habitat and to assess its condition. The focus is on the aquatic habitat for fish and macroinvertebrates. It is a physical assessment of the available habitat, rather than water quality or chemical aspects, or the specific requirements of particular species or communities.
The basic parameters included are:-
Many of the parameters have to be assessed visually. Often the sections may be turbid and hence it may be difficult to make the assessment. This can be noted and the types visible can be simply recorded 'presence only'.
Subjective Assessment
A subjective assessment is also made using a 5-level system ('Very Poor' to 'Very High') using the criteria provided on the datasheet.
A preliminary assessment of scenic, recreational and conservation values is made at each site. The assessment is very subjective, but nevertheless worthwhile in providing preliminary ratings in relation to other sites surveyed. The sites are classified according to their recreational opportunity type using remoteness, access, human contact and impact, and facilities available at the site. The suitability of the site for various types of recreation, both potential and current, is also assessed.
The sites are rated in terms of their overall scenic value (scored 1-10) and the attributes contributing to their scenic value are ranked.
A preliminary subjective assessment of the conservation values of the sites is made in terms of a rating score (1-10) for aquatic and riparian plants and animals and the value of the site as a wildlife corridor, using the criteria provided on the datasheet. Rare or endangered species known to occur in the area can be recorded. The sites are also rated as representative aquatic and riparian habitats of the types present at the site in relation to similar types surveyed elsewhere in the catchment.
This data sheet was designed to provide the basic data needed to classify urban streams in terms of channel modifications, reinforcements and buffer/setback features and dimensions. This classification defines the naturalness of the stream corridor, the type of modification present, and the limits or capability of the section for rehabilitation and restoration within the constraints imposed by the modifications. Streams with natural buffer and set-back zones, but with degraded bed and banks, have less value and capability for rehabilitation or restoration than streams that have not been channeled or modified in other ways. Likewise streams with natural channels but highly modified shoreline areas that have been cleared and converted to mown grass and pathways have their potential values limited by these modifications. Rehabilitating the stream by overcoming the degradation or installing artificial riffles or other habitat improvements will be ultimately limited by the modifications to the riparian zone and flood-plain areas. A bed, bank and buffer zone modification index has been developed. It is based on the three-zone buffer concept of Scheuler (1995) with the addition of the bed and 'bank slope' zones (Table 2).
The data collected include modifications to the bank shape and stream meander pattern through re-sectioning, realignment and various artificial linings and reinforcements. The bed and bank slope are classified as natural or modified and the percentage naturalness recorded as percentages of bed area and bank length in the survey reach. Artificial bed and bank protection methods are recorded. Other modifications in the three buffer zones are also recorded. An average weighted naturalness index for the corridor is produced from the five naturalness estimates. The bed, bank slope and shoreline zones have higher weightings. The average bankfull width, and the width of the bank slope and buffer zones are also recorded.
| Table 2. Zoning System Used to Classify Urban Streams Buffer zones (adapted from Scheuler 1995) | |||||
|---|---|---|---|---|---|
| Characteristics | Bed (instream) | Bank Slope | Streamside Buffer Zone | Middle Buffer Zone | Outer Buffer Zone |
| Function | Habitat | Habitat and maintain bedcondition and attributes | Protect the physical integrity of the stream ecosystem | Provide distance between the upland development and the streamside zone | Prevent encroachment and filter backyard runoff |
| Width | Dynamic or constrained by modifications | Dynamic or constrained by modifications | Minimum of 10m or bankfull width, plus wetlands and critical habitats | Minimum of 20-30m or 25 times bankfull width, 100 year floodplain width | Minimum of 10m setback to structures |
| Vegetative Target | Natural submerged and emergent vegetation | Natural diverse cover | Original native vegetation | Managed native vegetation, some clearing or thinning allowed | Grass with trees and other vegetation types preferred |
| Allowable Uses | Very restricted (e.g. flood control, footpaths etc.) | Restricted (e.g. some recreational uses, some stormwater controls, tree removal by permit) | Unrestricted | ||
This datasheet is optional. It provides a visual water quality assessment. There is also provision to record flow data from gauge records, and flow measurements can also be recorded for the cross sections. Water quality measurements can also be recorded. Although individual water quality measurements do not provide much information there are benefits in surveying a large number of sites over a relatively short period of time throughout the catchment. This may complement other water quality data collected at relatively few sites over a long period of time. Similar comments apply to flow measurements.
This datasheet is used to record estimates of sinuosity, channel gradient, and embeddedness (Rosgen 1996), and distance between pools. It was designed to generate a Rosgen type description of channel type. A subjective assessment of suitability for natural design is also made by rating the adequacy of the buffer width, the tenure of the buffer area, the links to good habitat upstream and the potential of the site for re-establishing plant and animal communities. Provision is also made for compiling non-survey data for natural design purposes.
This datasheet includes a set of questions designed to assess:
There is no subsection datasheet as the relevant information is entered directly into the database or is generated from information on the other datasheets and databases. This information includes the river and tributary names, the location of the subsection.
The length of major and minor streams in the section is also entered into the Sub-section database.
There is a separate database for each component and one for the subsections. There are a number of other databases for storing user options and for running the programs. The component databases mirror the datasheets so that the raw data is always available to the user. The 'trip-code' provides the linkage between the data for each site, basin, survey number and date combination. The ratings and various summary parameters are stored in the site and subsection databases after being calculated. Once the analyses have been completed both the raw and derived data are available to the users without the need for re-calculation.
In most instances the key issue is the extent to which the values or perceived function of an attribute has declined from a pristine or undisturbed condition. The ratings have been designed as percentages where 100% represents the full value, pristine condition or complete function for the component, and 0% represents a complete loss of these. Both absolute and relative standards are used. In-built formulae are used to derive percentage ratings compared with presumed pristine values. Comparisons with representative sites of similar type in good or pristine condition can be used to scale the ratings produced. The stability of the bed and banks is an example of an absolute standard. In assessing the condition of bed and banks we are interested in determining the extent to which the banks have become unstable through erosion, slumping, gullying, or the excessive build up of sediments on the bed or banks. Streams and rivers in good condition are assumed to be stable or in a graded state with some form of dynamic equilibrium prevailing between sediment supply, channel form and the sediment carrying capacity of the stream. The standard is 100% of bank length in stable condition. However some limited instability may be a characteristic of the type of stream even in the pristine state. In this case a site in pristine condition may generate an absolute score of 85% for example. This value is then scaled to 100% for the scaled rating. Both the absolute and scaled ratings are presented in the result summaries. The outputs are normally presented in 20% categories, and both the scaled and un-scaled ratings would have been allocated to the highest category (80-100%).
If no remnant sites in good condition are available, the presumed original pristine condition has to be derived from whatever data are available, including historical data, or use sites in other catchments or in the same general region. If no suitable reference sites are available it may be better to simply use the absolute rating with the maximum set at 100%. For management purposes the relative condition may be more important rather than the absolute values. The results are normally presented in categories and the category boundaries may be set to generate the required discrimination between sites on a relative scale. For example, two of the category options for the data summaries are:
| Rural Stream | Urban Stream |
| 0-20% | 0-10% |
| 21-40% | 10-25% |
| 41-60% | 26-40% |
| 61-80% | 41-60% |
| 81-100% | 61-100% |
This 'urban stream' categories effectively scales the ratings by placing emphasis on the sites with poorer condition. This may be more realistic in highly modified catchments. The other point is that the absolute rating is the more realistic when comparing catchments or sub-catchments. For example some streams may not store much organic debris or have a wide variety of instream cover even in a pristine state and the aquatic habitat scores may be less than 70% for example. It would be wrong to classify these sites as degraded because the 100% value was not achieved. On the other hand the value of their aquatic habitats may be less than in another stream where natural ratings were much higher. This highlights the difficulty in separating the relative absolute value of the components from their condition. References sites are an advantage, but they may be hard to find for every stream type especially in lowland flood plain areas. Absolute values may have to be used in urban situations where reference sites may not be available an where the absolute value of the component is more relevant as it reflects both the size of the original value and the extent of degradation.
The condition ratings are generated using fixed formulae from that data collected for each component. Modifying the formulae for urban streams was rejected because of the benefits in having a consistent and universal system. Scaling was selected as the best way to compare changes in condition between areas where the pristine absolute values were different. In all cases both the scaled and absolute values should be presented and the differences made explicit.
The ratings for the sections are calculated by using all the information that was collected from one or more sites within the sub-sections. The site ratings are averaged and the ratings re-calculated using the combined data set. Both of these are stored in the subsection database. Grouping fields are used to produce ratings and data summary parameters for groups of sites and subsections, for example for planning areas or for sub-catchments. The scaled ratings are also calculated and stored together with the original absolute values.
After being calculated the ratings for the sites (both scaled and un-scaled) are stored in the component databases. All the section ratings are stored in the section database. All the ratings can easily re-calculated and re-entered if required.
A variety of subjective and objective ratings are produced from the data collected using various formulae and computer programs. These ratings are produced for both the sites and sections. Various attributes and weightings have been applied in the formulae. The attributes are shown in the datasheets (Appendix 1). The weightings applied have been shown in 'bold' fonts in the formulae shown in the following sections.
Subjective Rating
This is simply the subjective rating converted to a percentage.
Calculated Rating - The calculated ratings are derived by weighting the various types of land tenure, local disturbance and land use in terms of their potential impact on the waterway and corridor. These are then applied using the formula.
| Calculated Rating = | tenure * (0.6) |
| + local disturbance * (0.15) (none=0.2) | |
| + land use *(0.2) |
For tenure the weightings were:
| 1 | Road Reserve, Urban, Unknown |
| 2 | Freehold, Leasehold |
| 3 | Urban Reserve |
| 4 | State Forest |
| 5 | Timber Reserve |
| 6 | State Park, Vacant Crown Land |
| 7 | for National Park |
| 1 | Urban Manufacturing/ Processing |
| 2 | Urban Residential, sugarcane |
| 3 | Horticulture tree crops/ fruit; Intensive livestock |
| 4 | Irrigated broad-acre crops |
| 5 | Non-irrigated broad-acre crops |
| 6 | Grazing-sown pasture |
| 7 | Grazing-native ---cleared; Horticulture- small crops and vines |
| 8 | Grazing-native ---thinned |
| 10 | Rural Residential/ Hobby farm |
| 14 | Grazing-native ---virgin timber |
| 15 | Park, Reserve |
For local disturbance the weightings were:
| 1 | Sand/Gravel mine |
| 2 | Other Mine |
| 3 | Dredging |
| 4 | Sewerage Effluent; Grazing |
| 5 | Sugar Mill |
| 6 | Forestry Activities |
| 10 | Irrigation Runoff, Pipe Outlet |
| 11 | Channelisation |
| 12 | River Improvement works |
| 13 | Water Extraction; Ford /Ramp |
| 14 | Bridge / Ford/ Culvert |
| 15 | Road |
The Channel Habitat Diversity Index - This is calculated by summing the natural logarithm of the cover estimates for each type present:-
| Diversity Index = |
| 10 * [ å ( Loge (percent for types present +1) ) - Loge (100) ] |
The value is set to 100 (minimum set to 1), and regarded as a percentage. This is similar to other diversity measures. The maximum value (100%) occurs when there are a wide variety of types present, each of which are present in equal percentages within the reach (e.g. 20% pool; 20% rapid; 20% cascade; 20% riffle; and 20% backwater). Reaches where only one type is present are scored as "1". Reaches dominated by one type such as 90% pool, also have low values for the index. This diversity index gives a measure of the diversity of channel habitat types and therefore the range of biological habitats present in the reach. Each channel habitat type has its own set of depths, flows, substrates and gradients.
Subjective Rating - The subjective rating is simply the re-scaled "Overall Instability Rating" and the "Susceptibility to Erosion Rating" combined for both banks. Scores of 100% occur when both banks are rated as having minimal instability and susceptibility to erosion.
| Subjective Rating = |
| (0.8) * Instability Rating + (0.2) * Susceptibility Rating |
Calculated Rating - The calculated rating is determined from the percentages of the lower and upper banks on both sides which are rated as "Stable" (recorded as 0-100%).
| Calculated Rating = |
| (0.8) * % Upper bank stable + (0.2) * % Lower bank stable |
The final rating is the average for the two banks.
Extra Rating - This is used to record the major direction of the instability, that is in terms of erosion (eroding and slumping) (<50%) or aggradation (>50%).
Subjective Rating - The subjective rating is simply the re-scaled "Overall Bed Stability Rating".
| Stable ("3") | => | (100%) |
| Moderate Erosion or Aggradation ( "2" or "4" ) | => | (50%) |
| Severe Erosion or Aggradation ( "1" or "5" ) | => | (20%) |
Calculated Rating - The index is derived by first converting the bed stability rating (BSR) to a percentage index value scaled for instability.
| BSR Rating | Stability | Index value |
| 1 | Highly eroding | 80 |
| 2 | Moderately eroding | 50 |
| 3 | Stable | 0 |
| 4 | Moderately aggrading | 50 |
| 5 | Highly aggrading | 80 |
The rating is calculated differently depending on whether the dominant process is aggradation or erosion, and also whether or not a bar is present.
If the beds are stable (BSR rating =3) and there is no bar present the bed is treated as eroding with the Index value set at 100. This value will be reduced slightly if there are many sources of potential instability present at the site (see 3.10.4.3).
The beds are aggrading if there is a bar present or the dominant process is aggrading (i.e. bar size >0 and/or bed stability rating >3)
Bed and Bar Rating = 100 - Index.
Index = [ Index + ( 1.5 * bar size) ] /2
Weightings for the type of bar present are then used as multipliers to adjust the derived indexes. Higher weights are used for the bar types indicative of a dominant process affecting much of the bed.
Index = Index * (bar type weight)
| Bar Type and Location | Weight |
| Point | 1.11 |
| Alternate | 1.11 |
| Island | 1.05 |
| Encroaching vegetation | 1.05 |
| Obstructions | 1.05 |
| Bar-plain | 1.18 |
| Infilled bed | 1.18 |
| High Flow deposits | 1.00 |
The index values are then scaled according to the angularity of the bed particles. The more angular the more recent the aggradation with input of new materials from outside the channel.
Index = Index * { ( 1 + (60) /1000) * ( 4 - gravel angularity category) }
The values are then scaled according to the shape of the bed particles. The more rounded the more mobile the particles and the more likely the sediment will continue to be passed downstream.
Index = Index * { ( 1 + (40) /1000) * ( 4 - gravel shape category) }
The index is then converted to a stability rating for the bed and banks:
Bed and Bar Rating = 100 - Index
If the beds are rated as stable or eroding (i.e. bed stability rating < 4) the index is based on the various recorded ratings for bed stability, sources of instability and bed controls. The index for the eroding beds are based on the BSR ratings and the percentage index values shown above. This index is then modified according to the factors controlling the stability, bed compaction, gravel shape and the controls present.
| Index = Index * { ( 1 + (50) /1000) * ( 5 - compaction category) } |
| (more compaction - less likely to erode) |
| Index = Index * { ( 1 + (40) /1000) * ( 4 - gravel shape category) } |
| (more rounded - more likely to erode) |
| Index = Index * | { ( 1 + (50) /100) * ( Log10 ( number of instability factors + 1) |
| - (60) /100) * ( Log10 ( number of control types + 1)} | |
| (more problems identified more likely to erode) | |
| (more controls present less likely to erode (negative effect)) | |
| Bed and Bar Rating = 100 | - (0.5 * Aggradation Index) |
| - (0.5 * Eroding Index) |
Extra Rating - This is used to record the type and severity of the instability.
Extra Rating = 50 - (Erosion index)/2 for erosion dominant
Extra Rating = 50 + (Aggradation index)/2 for aggradation dominant
The rating for riparian vegetation is based on the width of the riparian zone (50m wide rated as maximum) which is multiplied by a factor relating to the density and structural diversity of the vegetation in the riparian zone. In effect the width of the zone is taken as the original index (re-scaled to 100% for zones > 50m wide). This is then increased or decreased according to the diversity and density of the vegetation present. The maximum score of 100% is recorded for zones 40- 50m or greater in width, which have a good mixture of tall and low trees as well as a variety of understory species. Both overstory and understory are required for the riparian vegetation to perform all its functions.
The basic formula is:-
The width factor is derived using a quadratic equation to modify the actual recorded width. A non-linear function is used because the value of narrow zones is much less than the wider zones. The maximum riparian width is set to 50. The remnant riparian width ('ripwidth') is first doubled and then the following equation applied. The maximum riparian width is set to 50.
The vegetation factor is derived by assigning by grouping the vegetation into structural types, applying a suitable weight and then adding the group contributions. Each group contributes a maximum of 100 to the rating.
Cover reduction for Exotic Species
Each of the cover percentages are first reduced if exotic species are present.
| Modified cover (%) = | Original cover (%) - |
| 0.3* ( % exotic species) * Original cover (%). |
For example if the cover by tall trees is 70%, and half of the tall trees are exotic species the cover is reduced to 63% [i.e. (70 – (0.3*0.5) * 70)]. If all of the plants present are exotic species the cover is reduced by 30%.
The vegetation types are grouped into functional/structural types and assigned a weight related to their estimated relative importance in the riparian zone. Mangroves have the highest rating (1.0) and grasses the lowest (0.3). Bare ground has a negative weighting of 0.3. The total score for each group is set at 100.
| Weights | Structural/ functional Group (maximum contribution =100%) |
| +1.0 | Mangroves |
| +0.85 | Tall trees, medium trees |
| +0.70 | Small trees and palms |
| +0.55 | Shrubs, vines and tree ferns |
| +0.40 | Salt marsh plants, rushes, ferns, herbs |
| +0.30 | Grasses and mosses |
| - 0.30 | % of bare ground |
Adding the weighted scores for the group and dividing the result by 100 produces the vegetation factor.
| Vegetation factor = | [+ (1.00) * (mangroves ) |
| + (0.85) * (tall trees, medium trees) | |
| + (0.70) * (small trees and palms) | |
| + (0.55) * (shrubs, vines, tree ferns) | |
| + (0.40) * (salt marsh plants, rushes, ferns, herbs) | |
| + (0.30) * (grasses and mosses) | |
| - (0.30) * (% of bare ground)] /100 |
The maximum value for the vegetation factor is 2.8. Restricting the maximum contribution for each group to 100 ensures that the highest vegetation factors are produced when there is a diversity of structural vegetation types that are all relatively abundant in the zone. Tall or medium trees and shrubs will produce high values; grasses and herbs lower values. A mixture of tall trees and grasses will produce a maximum vegetation factor of 1.15, if both have 100% cover ratings. Grasses and ferns will produce a maximum factor of 0.7.
Multiplying the width and the vegetation factors produces the final rating.
Riparian Index = {width factor} * {vegetation factor} (expressed as a percentage)
The net effect of the formula is to weight the tall trees, mangroves more highly than the shrubs and reeds, but the maximum rating requires a variety of structural types to be present with relatively high cover. Narrow widths are rated poorly even if the vegetation cover within the narrow zone is high and diverse. Relatively wide zones of reed or grass, even when 50m wide are downgraded because of the absence of trees and shrubs. Allowing the vegetation factor to exceed 1.0 means that highly diverse and dense cover will boost the index above that expected for the zone width. The maximum score of 100% requires relatively wide zones, generally wider than 20m and an diverse range of vegetation types with relatively high cover scores.
The rating is derived by summing the weighted percentage covers for the various aquatic types present after adjusting for excessive cover and exotic species.
A 3rd order quadratic function is used to reduce the values of types where the recorded covers are in excess of 80% for some attributes. This is necessary because when the covers become too high they have a detrimental effect on habitat values. For example, a stream which is completely choked with water hyacinths has a degraded value compared with a stream that has only about 30-70% cover. The excessive plant cover may reduce light penetration, and may lead to periodic deoxygenation problems and eutrophication. Likewise a stream that is completely choked with reeds has a downgraded value as the reeds prevent passage of fish, they lead to accumulation of fine sediments and they reduce the diversity of habitat types present in the channel. The quadratic function has the effect of setting the maximum at 80% cover and thereafter reducing the index.
| Cover Index = (1.6057) | + (0.4443) * % Cover |
| + (0.04033) * % Cover2 | |
| - (0.00038875) * % Cover3 |
The cover indices for each type are then reduced by (0.3)* (% Exotics )*( cover index) for exotic species present for each type..
Weightings are then applied to produce the final percentage rating.
| Aquatic Vegetation Rating = | (0.5) * (cover index for submerged vegetation) |
| + | (0.3) * (cover index for emergent vegetation) |
| + | (0.2) * (cover index for floating vegetation) |
The relative weights reflect the estimated relative value of the various types as fish and macroinvertebrate habitat. The final rating will be higher if all three types are present and will have a maximum value less than 100% because of the adjustment for excessive vegetation cover. Low ratings do not necessarily imply a degraded habitat, they simply rate the potential value of the aquatic vegetation cover. The loss of canopy cover may allow aquatic vegetation to become established in areas where it was historically rare. However its value is related to its current presence and density not to its historic occurrence or absence.
i.e. Subjective Rating = ( 5 - Aquatic Habitat Rating) * 25
Calculated Rating - The basic formula is :-
Instream Cover – The components (logs, branches etc.) are first scaled using a 3rd order quadratic function because cover values in excess of about 80% represent degraded habitat.
| Cover Index = (1.6057) | + (0.4443) * % Cover |
| + (0.04033) * % Cover2 | |
| - (0.00038875) * % Cover3 |
A relative weight is then applied to the percentage cover for each type. These values are added.
| Cover Type | Weight | Cover type | Weight |
| Log | 0.90 | Submerged Veg. | 0.85 |
| Log jam <50% dense | 0.95 | Mangroves | 1.00 |
| Log jam >50% dense | 1.00 | Floating Veg. | 0.80 |
| Branch | 0.80 | Emergent Veg. | 0.80 |
| Branch pile <50% dense | 0.80 | Tree roots | 0.90 |
| Branch pile >50% dense | 0.85 | Rock surfaces | 0.75 |
| Leaf and twig | 0.40 | Pools >1m deep | 0.50 |
| Macrophyte debris | 0.50 | Man-made | 0.40 |
| Algae | 0.30 |
The final ratings are then re-scaled using a logarithmic function that tends to increase the rating given to small amounts of one type of habitat. However diverse types are required to generate the higher scores.
Instream Cover Index = (-24.791) + (54.820) Log 10 (Cover Index Total)
The maximum is set at 100.
Bank Cover - The bank covers are calculated for each side using the predicted maximum widths for each type to derive appropriate weighting for the recorded average width for the type. The weightings are adjusted according to their relative importance for the cover type.
| Bank Cover Type | Weight for width and relative importance |
| Canopy | 0.08 |
| Vegetation Overhang | 0.20 |
| Root Overhang | 0.25 |
| Bank Overhang | 0.40 |
| Man-made overhang | 0.05 |
The final contribution for each type is then
Width of type * percentage of bank length * weight for type
These individual ratings are then added together to give the final rating (max. = 100%).
| Bank Cover Index | = Canopy cover * 0.08 * average width |
| + Vegetation overhang * 0.20 * average width | |
| + Root overhang * 0.25 * average width | |
| + Bank cover * 0.40* average width | |
| + Man-made cover * 0.05 * average width |
The final ratings for the left and right banks are averaged and are then re-scaled using the same logarithmic function used for the instream cover. This increases the score to single types but a range of types are required to generate the higher scores.
Bank Cover Rating = (-24.791) + (54.820) Log 10 (Bank Cover Index)
The maximum score of 100% will generally require two or three types to be present along most of the bank. Canopy cover makes a large contribution if its average width is more than 10m.
The instream and bank cover indices are stored in the databases. A combined aquatic habitat rating is derived by weighting the two scores to derive a final rating expressed as a percentage:
Aquatic Habitat Index = (0.6) * (instream cover) + (0.4) * (bank cover).
This ratings are derived from the formulae shown below :
| Scenic / recreational value = | (0.6) * (recreational value) |
| + (0.4) * (scenic value) |
| Recreational Value | = {(0.7) * 10 *(10-recreational opportunity rating) |
| + (0.3) * 40 * {Loge (Number of recreational types listed)} |
where the contribution from the number of recreational types is set to 30.
Scenic Value = 10 * (Scenic Value Score)
The maximum is set to 100 and the ratings are expressed as percentages.
This rating is derived from the 5 ratings on the data sheet:
| Conservation Value = | 2 * ( Aquatic habitat rating + Riparian Habitat Rating + Wildlife Corridor Rating + Aquatic Representative Rating + Riparian Representative rating). |
This rating is expressed as a percentage.
Urban stream type indices are produced which include naturalness ratings for the bed and bank slopes, and naturalness and adequate width scores for the three buffer zones – streamside, middle and upper zones. The percent natural values are converted to a 6 level index:
| >95% | => | 6 |
| 76-95% | => | 5 |
| 51-75% | => | 4 |
| 26-50% | => | 3 |
| 6 | => | 2 |
| 1- 5% | => | 1 |
For the buffer zones these scores are adjusted for zone widths that are too narrow to be effective.
Stream side zone
If the average width <5m
Index set to 1
If the average width <10m and bankfull width >10m
Index set to 1
If the average width <10m and bankfull width 5-10m
Index set to 2
If the average width is less than twice the bankfull width
The index is reduced by 1.
Middle zone
If the average width <10m
Index set to 1
If the average width <30m and bankfull width >10m
Index set to 1
If the average width is 20-30m and bankfull width >10m
Index set to 2
If the average width is less than five times the bankfull width
The index is reduced by 1
Upper zone
If the average width <5m
Index set to 1
If the average width <10m and bankfull width >10m
Index set to 1
If the average width is <10m and bankfull width is 5-10m
Index set to 2
If the average width is less than twice the bankfull width
The index is reduced by 1
A combined rating is generated as a 5 figure number (e.g. 66665, 66454, 66223 etc.) by using each of the indices from bed to upper zone.
Average percent naturalness score
An average weighted naturalness score for the stream corridor is calculated using the following formula:
| Average % Naturalness rating |
= 0.40 * (bed naturalness)
+ 0.25 * (bank slope naturalness) + 0.20 * (streamside zone naturalness) + 0.10 * (middle zone naturalness) + 0.05 * (upper zone naturalness) |
No summary ratings are produced for water quality. The raw data and existing data summaries are used for assessment.
Rosgen Classification
A descriptive Rosgen type classification (Rosgen 1996) is derived from information on this data sheet and other data on the channel pattern (sheet 4) and the channel dimensions (sheet 6). This is only to be used as a descriptive classification of the type of channel at the site and not as a proper Rosgen classification. These classifications are labeled as 'n' and 'm' for natural and modified channels respectively. Sinuosity is not used to derive the classifications for the modified channels.
Suitability for Natural Design
This index is based on the average score for the four components on the data sheet:
where
| poor | => | 25% |
| moderate | => | 50% |
| good | => | 75% |
| very good | => | 100% |
Three ratings are produced from this data sheet – a constraints rating, and access rating and a rating the potential sources of water quality problems.
Constraints Rating
This rating is produced by summing the positive (+1) and negative (-1) answers to the questions. All the questions are equally weighted except the following, which were doubly weighed:
The results were expressed as a percentage of the maximum score of 22.
Access Rating
This was produced by adding the following values for each of the areas:
The result was expressed as a percentage of the maximum score of 640.
Each of the sources 'within the section' was allocated a score of 100, and the more 'remote sources' a score of 50. These score were then expressed as a percentage with the maximum set at 500.
The following data summary parameters are calculated for each section as well as the derived ratings. These parameters are also stored in the databases.
The software package contains a set of programs for checking and verifying the data and for calculating the derived ratings and scores.
The package provides a variety of data summaries and outputs.
The raw data is available in the databases. The derived ratings and summary parameters are also stored in the databases and can be easily accessed.
Comprehensive audit summaries of the condition of the stream corridor can be produced for sections and for user defined groups of sections. These summaries can be prepared for planning units, smaller tributary sub-catchments, and major sub-catchments and for entire catchments. Any combination of sections can be used. The user defines the groups using grouping fields in the database. These summaries can be generated as tabulated output (Appendix 2) or in graphical form as 'Stack Diagrams' (Appendix 3). These outputs include the various components and summary parameters listed in the previous section. Two category options are available for the percentage rating values.
| Option A | Option B | |
| 0-20% | 0-10% | |
| 21-40% | 11-25% | |
| 41-60% | 26-40% | |
| 61-80% | 41-60% | |
| 81-100% | 61-100% |
The results are presented as the percentage of the total length of major stream surveyed in the group of sections selected that have values within these category ranges. The tabulated output also shows length of stream placed into each category. The tabulated output provides additional information and shows the lengths of stream as well as the percentages.
An overall condition rating as a percentage score is produced which combines the component ratings. There are various choices for the weightings that can be applied for the various components. The option used for the summaries in Appendices 2-3 has a double weighting for riparian vegetation and single weightings for environs, bank, bed, scenic value, conservation value, aquatic vegetation.
These outputs provide a comprehensive overall assessment of the condition of the stream in various planning areas and sub-catchments. They define the severity and size of the problem in terms of the extent of the degradation. The outputs are also useful for comparing different areas or sub-catchments and also whole catchments.
The 'Report Card' (Appendix 4) provides a classification summary at the section and planning unit level. It provides a statistical summary of the condition ratings in terms of the number of sites present in the section or group of sections included in the summary. It also provides a statistical summary of the stream classification parameters and dimensions. The type of stream present, channel habitat types, bank shape, bank slope, and other bed and bank features are listed in order of their relative frequency in the sites. Likewise channel habitat types, instream cover, riparian vegetation types and bank cover types are similarly listed. Mean average sediment particle sizes are shown for each channel habitat type. Graphic summaries are also provided for the condition attributes, the urban stream classification codes (bed, bank and buffer zones) and dominant particle sizes.
These report cards provide a comprehensive summary of the classifications produced by the package for sections, planning units and other small groups of sites.
The package was designed to interface with GIS. A program is included that downloads the ratings and summary parameters for each section as Excel or text files. These outputs are designed for input to a GIS mapping system.
Unfortunately GIS is expensive to operate and not really available for community groups. It can be very frustrating waiting for GIS maps to be produced using the output data. An internal mapping system is included in the package that can produce simple 'skeleton' maps as screen or hard copy output. Samples of these skeleton maps are shown in Appendix 5. Each of the sections has a uniform length on these maps and the maps are greatly simplified sketch maps for the area. The user of the program produces the maps. Section numbers and key location points and other label can be incorporated onto the maps.
Each of the condition components and all of the summary parameters can be selected and displayed in various colour-coded categories using these maps. Maps from pilot studies in Norman Creek are shown in Appendix 5. These show overall condition, riparian vegetation condition, bank condition, bed and bar condition, aquatic habitat, conservation, bed and corridor naturalness rating. An additional map shows the sediment particle size for riffles. All of the condition parameters were displayed in five categories (0-10%; 10-25%, 25-40%, 41-60% and 61-100%). These maps show the location of the classified stream sections. They can be used to identify where major problems occur and where to start rehabilitation or other management activities.
A query language is included that allows a large range of combinations of the summary parameters to be displayed. A set of examples is provided for the following targets for Norman Creek (one of the pilot study catchments):
Rehabilitation of Aquatic Habitat
| Rating | Weight Applied | Comment |
| Average naturalness | 200 | Bed, bank and buffer naturalness is high |
| Riparian vegetation | 200 | Good riparian vegetation |
| Aquatic habitat | -50 | Selects poor existing habitat |
| Channel habitat diversity | -100 | Low existing diversity |
| Suit natural design | 100 | Suit natural design |
| Constraints | -50 | Few constraints |
Rehabilitation of Riparian Vegetation
| Rating | Weight Applied | Comment |
| Bank condition | 200 | Banks stable |
| Average naturalness | 200 | Bed, banks and buffer zones not highly modified |
| Riparian vegetation | -100 | Low existing vegetation |
| Constraints | -100 | Few constraints |
| Selection Criteria | ||
| Bed, bank and buffer naturalness index | >40000 | Selects sites with more than 70% of the bed in natural state |
Good Fish Habitat
| Rating | Weight Applied | Comment |
| Aquatic Habitat | 100 | Good range of bank and instream cover |
| Average naturalness | 100 | Naturalness high |
| Riparian vegetation | 100 | Good riparian vegetation |
| Channel habitat diversity | -100 | High channel habitat diversity |
| Selection Criteria | ||
| Pools depths at the 'water mark' | >0.5m | Selects site with pools present, and pool depth >0.5m |
The negative weightings were applied to select low values for channel habitat diversity and for constraints. Maps for these ratings, using eight categories, are shown in Appendix 5.
Comprehensive statistical summaries can be produced which include single components or all components. These summaries provide output for all the raw data included in the surveys and also the derive ratings. A sample output for the entire Norman Creek Catchment is shown in Appendix 6. One aspect of the output needs to be clarified. The statistics are presented as mean cover or bank length estimates only for sites where the attribute is present. Zero values are not included when calculating the statistics. The outputs also show the number of sites where the attribute was present. These outputs usually summarized as the percentage of sites where the attribute was found and then the average cover or length of bank for these sites.
The classification system provides a comprehensive system for managing urban streams. The two major output types – audit summary and map displays provide the key types of information needed for management:
How severe are the problems? How do my streams compare with those in adjacent catchments?
The audit summaries list the percentage of the total stream length that is classified into various condition categories. This quantifies the size and severity of the problem either in absolute terms or by comparison with other catchments or sub-catchments. The use of a single standard method for urban and rural catchments allow for realistic and consistent comparisons.
How large is the problem? What length of stream is affected to what degree or extent?
The audit summaries provide information on the length of stream affected in various ways and to various extents. This information is also presented as percentages for comparing catchments.
Where are the problems located? Where are the best sites to begin the rehabilitation? Where would the efforts of 'Bush Care' or some other interest group be most effective?
The map outputs and the data downloaded to GIS show the locations of selected areas using colour-coded ratings of relative severity of a problem or relative suitability of an area for various purposes. The user can develop their own tests by combining the derived ratings and summary parameters. The database system can be used to move down through the data hierarchy to explore the derivation of the various ratings. Most of the survey data can also be plotted. Likewise the data added to the databases from existing sources can also be plotted. This system provides a very powerful and flexible system for section selection and establishing priorities.
To re-establish fish populations I need to find areas with permanent pools at least 1m deep, with abundant aquatic vegetation and good instream cover in the form of logs and branches. Where do I find such habitat areas?
The query system and the mapping package or the GIS system can locate sections that meet these criteria. The user can define a formula that locates and ranks the suitability of various sections as aquatic habitat.
Where in the catchment are there representative sites of various types that will provide a reasonable target for rehabilitation or management initiatives? What areas of riparian vegetation and stream habitats should be conserved or protected?
The mapping package will show the location of the best representatives for each component. The query system can be used to further refine the selection and rating of the sites. For example the user may want to select sites that have both riparian vegetation and aquatic habitats in good condition. Likewise, the user may want to select sites that have good riparian vegetation and natural bed and banks. The use of derived and raw survey data provides a wide variety of options for narrowing the search and increasing the power of the selection to specifically meet the requirements.
The surveys have been designed to be repeated at regular intervals to assess the trends and rates of change and dominant processes. Follow-up surveys can also be used to monitor the outcomes of various management initiatives or rehabilitation strategies. It is suggested that the surveys be repeated every 3-5 years. The rapid survey approach and the relatively low cost mean that this monitoring is feasible. The software system is also designed for multiple surveys.
The accuracy and reliability of the data collected ultimately limits the accuracy and reliability of the method. Using non-experts to conduct the surveys places obvious limits on the type of data that can be collected and the consistency and reliability of the data. These limitations are offset by the major advantages of involving community groups, the low cost and the short time for the surveys. It is feasible to complete surveys of 200 sites and to completely establish the software system in three months (Anderson 1999b). Involvement of local community groups in planning and undertaking the surveys is vital for creating a sense of local ownership of the data and the outputs. The software system has been designed to be operated by community groups on a laptop computer. The 'Skeleton' mapping system is limited to preparing preliminary maps, but again it provides output to those who may not have access to GIS.
The limitations and subjectivity of some of the assessment criteria mean that some inconsistency is to be expected. The ability to move through the hierarchy of raw and derived data may overcome some of these problems. Likewise the data collected is very comprehensive and so some of these problems may be overcome by examining other related data. The system is designed to deal with complex issues to select, rank and establish priorities between the various sections of waterway in a preliminary way. It does not provide all the answers. The next stage is to conduct detailed investigations of the sections nominated. The system is therefore designed as an exploratory tool rather than a comprehensive management system. It is a 'Basic Decision Support System' rather than a model-based high-level DSS.
The absence of biological information or water quality also imposes other limitations. However the system has been designed to interface with this information, provided it can be organised to mesh with the sampling regime associated with the 'homogeneous stream sections'. The heart of the system is a database and consequently the data it contains can be readily and easily linked with other database information through GIS or through other software systems. The only impediment for these transfers is the sampling design. Provided data can be collected for the defined 'homogeneous' sections there are major opportunities to link the data with all sorts of information not only for the waterways and corridors but also to the land area within the catchment. The basic sampling element of the stream section and its immediate local sub-catchment or drainage area is normally applied to the whole catchment. This provides a link to the catchment land surfaces. Thus there is a simple way of interfacing with all sorts of stream and land area data.
Pilot studies of the methodology have been conducted in six catchments in Brisbane with the cooperation of Brisbane City Council (Anderson (1999b). The software system was also trialed by Brisbane City Council and community groups. The extra resources provided by Brisbane City council meant that the pilot study was extended to a major complete exercise for the six catchments.
Audit Summary Part A
|
Audit Summary Part B
|
Habitat Summary Part A
|
Habitat Summary Part B
|
Habitat Summary Part C
|
(intentionally blank) |
Report Card A4-A
|
Report Card A4-B
|
Report Card A4-C
|
Report Card A4-D
|
Report Card A4-E
|
(intentionally blank)
|
'Skeleton' Map A5-A: Overall Condition
|
'Skeleton' Map A5-B: Bank Stability
|
'Skeleton' Map A5-C: Bed & Bars
|
'Skeleton' Map A5-D: Riparian Veg
|
'Skeleton' Map A5-E: Aquatic Habitat
|
'Skeleton' Map A5-F: Conservation Value
|
'Skeleton' Map A5-G: Modification Rating
|
'Skeleton' Map A5-H: Rehab' Aquatic Habitat
|
'Skeleton' Map A5-I: Rehab' Riparian Veg
|
'Skeleton' Map A5-J: Fish Habitat
|
'Skeleton' Map A5-K: Riffles
|
(intentionally blank) |
Statistical Summary: Channel Habitat
|
Statistical Summary: Environs
|
Statistical Summary: Environs
|
Statistical Summary: Channel Dimensions All Types
|
Statistical Summary: Channel Dimensions Pools
|
Statistical Summary: Channel Dimensions Non Pools
|
Statistical Summary: Channel Dimensions Riffles
|
Statistical Summary: Channel Dimensions Runs
|
Statistical Summary: Channel Dimensions Glides
|
Statistical Summary: Channel Dimensions Cascades
|
Statistical Summary: Channel Dimensions Rapids
|
Statistical Summary: Bank Report
|
Statistical Summary: Beds & Bars
|
Statistical Summary: Vegetation
|
Statistical Summary: Vegetation
|
Statistical Summary: Aquatic Habitat
|
Statistical Summary: Scenic & conservation
|
Statistical Summary: Scenic & conservation
|
Statistical Summary: Channel Features
|
Statistical Summary: Channel Features
|
Statistical Summary: Waterway Class
|
Statistical Summary: Constraints & Opportunities
|
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