Previous Page Table of Contents Next Page


Environmental Flow Assessment With Emphasis on Holistic Methodologies

Arthington A.H.1, 2 Tharme R.E.3 Brizga S.O.4 Pusey B.J.1, 2 Kennard M.J.1

1 Co-operative Research Centre for Freshwater Ecology, Centre for Riverine Landscapes, Faculty of Environmental Sciences, Griffith University, Nathan, Queensland, Australia
2 Co-operative Research Centre for Tropical Rainforest Ecology and Management, Centre for Riverine Landscapes, Faculty of Environmental Sciences, Griffith University, Nathan, Queensland, Australia
3 International Water Management Institute (IWMI), P.O. Box 2075, Colombo, Sri Lanka
4 S. Brizga & Associates Pty Ltd, PO Box 68, Clifton Hill, Victoria

Address for correspondence: A.H. Arthington, Centre for Riverine Landscapes, Faculty of Environmental Sciences, Griffith University, Nathan, Queensland, Australia E-mail: a.arthington@mailbox.gu.edu.au

Key words: Environmental Flow Assessment, holistic methodologies, rivers, floodplains, fisheries models, flow experiments, research priorities

ABSTRACT

Worldwide there is growing awareness of the pivotal role of the flow regime (hydrology) as a key ‘driver’ of the ecology of rivers and their associated floodplain wetlands. Ecological processes related to flow and other factors govern the ecosystem goods and services that rivers provide to humans, such as flood attenuation, water purification, production of fish and other foods and marketable goods. Protecting and restoring river flow regimes and hence the ecosystems they support by providing environmental flows has become a major aspect of river basin management. Over 200 approaches for determining environmental flows now exist and they are used or proposed for use in more than 50 countries worldwide. Most methodologies currently used in Australia and southern Africa and increasingly in other countries, are holistic in their scope, recognising that it is necessary to provide water for aquatic ecosystems from source to sea and for all water-dependent ecological components. This paper provides a brief history of the development of environmental flow methods and identifies the main features and strengths of each, giving most emphasis to holistic or ecosystem methodologies. We then present an overview of research initiatives needed to enhance these approaches and improve their capacity to predict the ecological, social and economic consequences of change in river flow regimes.

INTRODUCTION

In many parts of the world there is growing awareness of the pivotal role of the flow regime (hydrology) as a key ‘driver’ of the ecology of rivers and their associated floodplains (see Richter et al. 1996; Poff et al. 1997; Puckridge et al. 1998; Bunn and Arthington 2002; Naiman et al. 2002 for reviews). Every river system has an individual or ‘signature’ flow regime with particular characteristics relating to flow quantity and temporal attributes such as seasonal pattern of flows, the timing, frequency, predictability and duration of extreme events (e.g. floods and droughts), rates of change and other aspects of flow variability (Richter et al. 1996; Poff et al. 1997; Olden and Poff 2002). Each of these hydrological characteristics has individual as well as interactive regulatory influences on the biophysical structure and functioning of river and floodplain ecosystems, including the physical nature of river channels, sediment regime and water quality, biological diversity/riverine biota and key ecological processes sustaining the aquatic ecosystem (Naiman et al. 2002). These processes in turn govern the ecosystem goods and services that rivers provide to humans (e.g. flood attenuation, water purification, production of fish and other foods and marketable goods).

In large part, recognition of the importance of flow and its interactions with other driving variables has stemmed from an increasing body of information describing the negative impacts to riverine ecosystems that are clearly attributable, either directly or indirectly, to the alteration of natural flow regimes (Rosenberg, McCully and Pringle 2000; Bunn and Arthington 2002).

Recognition of the escalating hydrological alteration of rivers on a global scale and the resultant environmental degradation has led to the gradual establishment of a field of scientific research termed environmental flow assessment (EFA) (Tharme 2003). In simple terms, such an assessment addresses how much and which specific temporal characteristics, of the original flow regime of a river should continue to flow down it and onto its floodplains in order to maintain specified features of the riverine ecosystem (Arthington et al. 1992; Tharme and King 1998; King, Tharme and De Villiers 2002). An EFA produces one or more descriptions of possible modified hydrological regimes for the river, the environmental flow requirement(s) (EFRs) or environmental water allocation(s), each regime linked to a predetermined objective in terms of the ecosystem’s future condition.

Environmental flow assessments are directed at two main types of management response to the potential and extant impacts of altered flow regimes:

(1) A proactive response, intended to maintain the hydrological regimes of undeveloped rivers as close as possible to the un-regulated condition, or at least to offer some level of protection of natural river flows and ecosystem characteristics, and (2) A reactive response, intended to restore certain characteristics of the pre-regulation flow regime and ecosystem in developed rivers with modified/regulated flow regimes. Both of these circumstances can be addressed using the environmental flow assessment methods currently available.

The level of resolution of the EFR produced may range from a single annual flow volume through to, more commonly nowadays, a comprehensive, modified flow regime where the overall volume of water allocated for environmental purposes is a combination of different monthly and more frequent, event-based flow quantities such as within-channel or floodplain flood pulses (Tharme 2003). The scale at which the assessment is undertaken may also vary widely, for instance, from an entire large river basin that includes a regulated main channel and/or several regulated tributaries, to a flow restoration project for a single flow-impacted river reach (Arthington, Brizga and Kennard 1998; King, Tharme and Brown 1999). Different methodologies are appropriate at each particular spatial scale as well as in relation to typical project constraints, including the time frame for assessment, the availability of data, technical capacity and finances (Tharme 1996; Arthington et al. 1998; Arthington, et al. 2003a; Kennard et al. 2003b). Methodologies accordingly range from rapid, reconnaissance-level approaches for regional, national or basin wide water resources planning, to resource intensive methodologies for highly exploited, individual river sites subject to multiple uses or rivers of high conservation significance.

This paper describes the origins of methods for environmental flow assessment and the different types of approaches presently available. It does not describe individual methods in detail, as many reviews, case studies and manuals are available (inter alia Bovee 1982, 1998; Milhous, Updike and Schneider 1989; Arthington and Pusey 1993; Stalnaker et al.; Tharme 1996; Jowett 1997; Stewardson and Gippel 1997; Dunbar et al. 1998; Arthington 1998; Arthington and Zalucki 1998; Dunbar et al. 1998; Milhous 1998; King et al. 2002, 2003, 1999; Tharme 2003). Further information on recent world developments in the field of environmental flow assessment can be found in River Research and Applications Volume 19 (2003) containing selected papers from the International Working Conference on Environmental Flows for River Systems and the Fourth International Ecohydraulics Symposium (held in Cape Town, South Africa, March 2002).

The main focus of this paper is the category of techniques termed holistic methodologies (sensu Tharme 1996) and their diversification to address both river ecosystem protection (i.e. proactive approaches) and river ecosystem restoration (i.e. reactive approaches). We outline the characteristics, strengths and limitations of the main holistic methodologies in use today and comment on shared features and best practices that commend these methods for use in developed as well as developing countries. We conclude with a brief overview of the modelling and research initiatives needed to enhance these holistic methodologies and increase their capacity to produce quantitative predictions of the effects of altering a river’s flow regime outcomes and eventually, predictive models of the ecological and knock-on socio-economic consequences of changes in river flow regimes.

THE ORIGINS OF ENVIRONMENTAL FLOW ASSESSMENT

Tharme (1996) traced the evolution of environmental flow methodologies worldwide, observing that historically, the United States of America was at the forefront of research with the first ad hoc methods appearing in the late 1940s and a series of more formally documented techniques emerging in the late 1970s. In most other parts of the world, EFA processes became established far later, with approaches to determine environmental water allocations only beginning to appear in the literature in the 1980s. Early on and still today in some countries, the focus of environmental flow assessment was the maintenance of economically important freshwater fisheries, especially salmonid fisheries, in regulated rivers. The main objective was to define a minimum acceptable flow based almost entirely on predictions of instream habitat availability matched against the habitat preferences of one or a few species of fish (see Jowett 1997; Pusey 1998 for reviews). It was assumed that the flows recommended to protect target fish populations, habitats and food resources would ensure maintenance of the river ecosystem. From these early attempts to quantify appropriate stream flows for fish, many new methods and innovations have evolved and recently, a much more comprehensive approach to EFAs has been adopted in both theory and practice.

DESCRIPTION OF METHODOLOGIES

Tharme (1996, 2003) has recognised four relatively discrete types of environmental flow methodology: (1) hydrological, (2) hydraulic rating, (3) habitat simulation and (4) holistic methodologies; among other techniques occasionally applied during EFAs. The four types are briefly described below.

HYDROLOGICAL METHODOLOGIES

These represent the simplest set of techniques where, at a desktop level, hydrological data, as naturalised, historical monthly or average daily flow records, are analysed to derive standard flow indices which then become the recommended environmental flows. Commonly, the EFR is represented as a proportion of flow (often termed the ‘minimum flow’, e.g. Q95 - the flow equalled or exceeded 95 percent of the time) intended to maintain river health, fisheries or other highlighted ecological features at some acceptable level, usually on an annual, seasonal or monthly basis. In a few instances, secondary criteria in the form of catchment variables, hydraulic, biological or geomorphological parameters are also incorporated. As a result of the rapid and non-resource intensive provision of low resolution flow estimates, hydrological methodologies are generally used mainly at the planning stage of water resource developments, or in situations where preliminary flow targets and exploratory water allocation trade-offs are required (Tharme 1996; Arthington et al. 1998; Tharme 2003).

HYDRAULIC RATING METHODOLOGIES

Hydraulic rating methodologies use changes in simple hydraulic variables, such as wetted perimeter or maximum depth, usually measured across single, flow-limited river cross-sections (commonly riffles), as a surrogate for habitat factors known or assumed to be limiting to target biota. Environmental flows are determined from a plot of the hydraulic variable(s) against discharge, commonly by identifying curve breakpoints where significant percentage reductions in habitat quality occur with decreases in discharge. It is assumed that ensuring some threshold value of the selected hydraulic parameter at a particular level of altered flow will maintain aquatic biota and thus, ecosystem integrity. These relatively low-resolution hydraulic techniques have been superseded by more advanced habitat modelling tools, or assimilated into holistic methodologies (Tharme 1996; Jowett 1997; Arthington and Zalucki 1998; Tharme 2003). However, select approaches continue to be applied and evaluated, notably the Wetted Perimeter Method (e.g. Gippel and Stewardson 1998).

HABITAT SIMULATION OR MICROHABITAT MODELLING METHODOLOGIES

Habitat simulation methodologies also make use of hydraulic habitat-discharge relationships, but provide more detailed, modelled analyses of both the quantity and suitability of the physical river habitat for the target biota. Thus, environmental flow recommendations are based on the integration of hydrological, hydraulic and biological response data. Flow-related changes in physical microhabitat are modelled in various hydraulic programs, typically using data on depth, velocity, substratum composition and cover; and more recently, complex hydraulic indices (e.g. benthic shear stress), collected at multiple cross-sections within each representative river reach. Simulated information on available habitat is linked with seasonal information on the range of habitat conditions used by target fish or invertebrate species (or life-history stages, assemblages and/or activities), commonly using habitat suitability index curves (e.g. Groshens and Orth 1994). The resultant outputs, in the form of habitat-discharge curves for specific biota, or extended as habitat time and exceedence series, are used to derive optimum environmental flows. The habitat simulation-modelling package PHABSIM (Bovee 1982, 1998; Milhous 1998, 1982; Milhous et al. 1989; Stalnaker et al. 1994), housed within the Instream Flow Incremental Methodology (IFIM), is the pre-eminent modeling platform of this type. The relative strengths and limitations of such methodologies are described in King and Tharme (1994); Tharme (1996); Arthington and Zalucki (1998); Pusey (1998) and they are compared with the other types of approach in Tharme (2003).

HOLISTIC METHODOLOGIES

Over the past decade, river ecologists have increasingly made the case for a broader approach to the definition of environmental flows to sustain and conserve river ecosystems, rather than focusing on just a few target fish species (Arthington and Pusey 1993; King and Tharme 1994; Sparks 1992, 1995; Richter et al. 1996; Poff et al. 1997). From the conceptual foundations of a holistic ecosystem approach (proposed by Arthington et al. 1992), a wide range of holistic methodologies has been developed and applied, initially in Australia and South Africa and more recently in the United Kingdom. This type of approach reasons that if certain features of the natural hydrological regime can be identified and adequately incorporated into a modified flow regime, then, all other things being equal, the extant biota and functional integrity of the ecosystem should be maintained (Arthington et al. 1992; King and Tharme 1994). Likewise, Sparks (1992, 1995) suggested that rather than optimising water regimes for one or a few species, a better approach is to try to approximate the natural flow regime that maintained the "entire panoply of species".

Importantly, holistic methodologies aim to address the water requirements of the entire "riverine ecosystem" (Arthington et al. 1992) rather than the needs of only a few taxa (usually fish or invertebrates). These methodologies are underpinned by the concept of the "natural flows paradigm" (Poff et al. 1997) and basic principles guiding river corridor restoration (Ward et al. 2001,; Uehlinger et al. 2001). They share a common objective - to maintain or restore the flow-related biophysical components and ecological processes of in-stream and groundwater systems, floodplains and downstream receiving waters (e.g. terminal lakes and wetlands, estuaries and near-shore marine ecosystems).

Ecosystem components that are commonly considered in holistic assessments include geomorphology, hydraulic habitat, water quality, riparian and aquatic vegetation, macroinvertebrates, fish and other vertebrates with some dependency upon the river/riparian ecosystem (i.e. amphibians, reptiles, birds, mammals). Each of these components can be evaluated using a range of field and desktop techniques (see Tharme 1996; Arthington and Zalucki 1998; Tharme 2003; for reviews) and their flow requirements are then incorporated into EFA recommendations, using various systematic approaches as discussed in more detail below.

Holistic environmental flow assessments may include evaluation of a range of other mitigation measures, for example, how to restore longitudinal and lateral connectivity by providing fish passes or altering the configuration of levee banks on a floodplain. Management of storage water levels may also be examined and recommendations made on the benefits of more, or less, stable water levels. Some holistic methodologies also take into consideration the influence of threatening processes and disturbances unrelated (or less directly related) to flow regulation and advise on possible mitigation measures such as riparian and habitat restoration, or the management of invasive vegetation and fish.

TYPES OF HOLISTIC METHODOLOGY

Holistic methodologies currently represent around 8 percent of the global total, with at least 16 extant methodologies based on the holistic principles described above having been developed over the last ten years (Tharme 2003). Although predominantly developed and used in South Africa and Australia, recently such methods have begun to attract growing international interest in both developed and developing regions of the world, with strong expressions of interest from in excess of 12 countries in Europe, Latin America, Asia and Africa (Tharme 2003).

These approaches have been described (see Arthington et al. 1998) as either ‘bottom-up’ methods (designed to ‘construct’ a modified flow regime by adding flow components to a baseline of zero flows), or ‘top-down’ methods (addressing the question, "How much can we modify a river’s flow regime before the aquatic ecosystem begins to noticeably change or becomes seriously degraded?").

For comparative purposes, selected holistic methodologies are summarised in Table 1, in terms of their origins, key features, strengths, limitations and present stage of development and application (adapted from Tharme 2003). Further details of the various methodologies are available in the source references provided in Table 1, as well as in the review papers listed herein.

The South African Building Block Methodology or BBM (King and Tharme 1994; King and Louw 1998; King et al. 2002) was the first structured approach of this type. It began as a bottom-up method, more recently incorporating the Flow Stress- Response Method (O’Keeffe and Hughes 2002). In this modified form, the BBM is legally required for intermediate and comprehensive determinations of the South African Ecological Reserve (DWAF 1999). Other essentially bottom-up methodologies include ‘expert’ and ‘scientific panel’ methods developed and applied in Australia (reviewed in Cottingham, Thoms and Quinn 2002).

There are several so-called ‘top-down’ methods. Examples of top-down methods are the Benchmarking Methodology (Brizga et al. 2001) used routinely in Queensland (Australia) at the planning stage of new developments to assess the environmental impacts likely to result from future water resource developments and DRIFT - Downstream Response to Imposed Flow Transformations (King, Brown and Sabet 2003), a scenario-based approach that also predicts the probable ecological impacts of various scenarios of flow regime change. These methodologies make such predictions in different ways, as outlined in Table 1 and the background literature cited for each method therein. The Flow Restoration Methodology (Arthington et al. 1999; Arthington et al. 2000) is a bottom-up approach with the objective of shifting a regulated flow regime and river system more towards its natural state, combined with a simple top-down appraisal of the probable ecological consequences of not restoring certain features of the pre-regulation flow regime. The Flow Events Method (Stewardson and Cottingham 2002) seems to be a rather similar approach, usually linked to a scientific panel method (Table 1).

Table 1: Summary of holistic environmental flow methodologies presented in approximate sequence of development, highlighting salient features, strengths and limitations, as well as their current status in terms of development and application (adapted from Tharme 2003). Further information on the strengths and deficiencies of individual holistic methodologies is provided in Tharme (1996); Arthington (1998); Cottingham et al. (2002); Arthington et al. (2003a); King et al. (2003); Tharme (2003). Abbreviations: DNR - Queensland Department of Natural Resources; DWAF - South African Department of Water Affairs and Forestry; EAFR - ecologically acceptable flow regime; EF - environmental flow; EFA - EF assessment; EFR(s) - EF requirement(s); EFM - EF methodology; TAP - technical advisory panel; WAMP - water allocation and management planning; WRD(s) - water resource development(s); abbreviations for methodology names are given in the first table column.

Methodology

Origins

Features, strengths and limitations

Status

Holistic Approach:
(Arthington et al. 1992; Davies et al. 1996; Arthington 1998; Petit et al. 2001).

Developed in Australia to address EFRs of entire riverine ecosystem; shared conceptual basis with BBM and the theoretical and conceptual basis of the Benchmarking Methodology and Flow Restoration Methodology.

Conceptual and theoretical approach for bottom-up construction of EF regime for whole riverine ecosystem from headwaters to floodplains, including groundwater and estuary or coastal waters; describes systematic construction of a modified flow regime, on a month-by-month (or shorter time scale) flow element-by-element basis and based on best available scientific data, to achieve predetermined objectives for future river condition of rivers; principally rep-resents a flexible conceptual framework, elements of which have been adapted in a variety of ways into several Australian holistic methodologies and for individual studies; basic tenets and assumptions as per BBM, which was derived from it; incorporates more detailed assessment of flow variability than early BBM studies; includes method for generating trade-off curves for examining alternative water use scenarios; some risk of inadvertent omission of critical flow events (common to all holistic methodologies); applicable to regulated or unregulated rivers and for flow restoration; high potential for application to other aquatic ecosystems; recommends a monitoring programme as a crucial component of holistic flow assessments; lack of structured set of procedures and clear identity for EFM hinders rigorous routine application (but routinely used in customized format in Western Australia).

Represents conceptual and theoretical basis of most other holistic EFMs; developed and applied in various forms in Australia, e.g. expert panel assessments, Flow Events Method, Benchmarking Methodology and Flow Restoration Methodology

Building Block Methodology (BBM):
(King and Louw 1998; King et al. 2000).

Developed in South Africa by local researchers and DWAF, through application in numerous water resource development projects to address EFRs for entire riverine ecosystems under conditions of variable resources; adapted for intermediate and comprehensive determinations of the ecological Reserve under the new SA Water Law.

Rigorous and extensively documented (manual and case studies available); prescriptive bottom-up approach with interactive scenario development; moderate to highly resource intensive; shared conceptual basis with Holistic Approach; developed to differing extents for both intermediate-level (2 months) or comprehensive (1-2 years) EFAs, within South Africa’s Reserve framework; based on a number of sites within representative and/or critical river reaches; includes a well established social component (dependent livelihoods); functions in data poor or rich situations; comprises 3-phase approach: (1) preparation for workshop, including stakeholder consultation, desktop and field studies for site selection, geomorphological reach analysis, river habitat integrity and social surveys, objectives setting for future river condition, assessment of river importance and ecological condition, hydrological and hydraulic analyses, (2) multidisciplinary workshop-based construction of modified flow regime through identification of ecologically essential flow features on a month-by-month (or shorter time scale), flow element-by-flow element basis, for maintenance and drought years, based on best available scientific data, (3) linking of EFR with water resource development engineering phase, through scenario modelling and hydrological yield analysis; EFM exhibits limited potential for examination of alternative scenarios relative to DRIFT, as BBM EF regime is designed to achieve a specific predefined river condition; incorporates a monitoring programme and additional research on important issues, as crucial components of EF implementation; some risk of inadvertent omission of critical flow events (common to all holistic methodologies), high potential for application to other aquatic ecosystems; links to external stakeholder and public participation processes; flexible and amenable to simplification for more rapid assessments; less time, cost and resource intensive than DRIFT; applicable to regulated or unregulated rivers and in flow restoration context; now incorporates Flow Stressor-Response Method facilitating top-down, scenario-based assessments of alternative flow regimes, each with expression of the potential risk of change in river ecological condition.

Most frequently used holistic EFM globally, applied in 3 countries; adopted as the standard EFM for South African Reserve determinations

Expert Panel Assessment Method (EPAM):
(Swales and Harris 1995).

First multidisciplinary panel based EFM used in Australia, developed jointly by the New South Wales Departments of Fisheries and Water Resources.

Bottom-up, reconnaissance-level approach for initial assessment of proposed WRDs with many conceptual features and methodological procedures in common with the Holistic Approach and BBM; rapid and inexpensive, with limited field data collection; site-specific focus; applicable primarily for sites where dam releases are possible; relies on field-based ecological interpretation, by a panel of experts, of different multiple trial flow releases (ranked in terms of scored ecological suitability) from dams, at one or a few sites, to determine EFR (typically expressed as flow percentiles); low resource intensity; limited resolution of EF output; aims to address river ecosystem health (using fish communities as indicators), rather than to assess multiple ecosystem components; strongly reliant on professional judgement; limited subset of expertise represented by panel (e.g. fish, invertebrates, geomorphology); simplistic in terms of the range of ecological criteria and components assessed (but scope for inclusion of additional ones) and the focus on fish; no explicit guidelines for application; poor congruence in opinion of different panel members (e.g. due to subjective scoring approach, individual bias); requires further validation; led to development of more advanced, but similar SPAM, Snowy Inquiry Methodology and other expert panel approaches.

Applied only in Australia; several applications, both in original and variously modified forms

Scientific Panel Assessment Method (SPAM):
(Thoms et al. 1996; Cottingham et al. 2002).

Developed during an EFA for the Barwon-Darling River System, Australia.

Bottom-up field (multiple sites) and desktop approach appropriate for provision of interim or intermediate level EFAs with many conceptual features and methodological procedures in common with the Holistic Approach and BBM; evolved from EPAM as more sophisticated and transparent expert-panel approach; aims to determine a modified flow regime that will maintain ecosystem health; differs from EPAM in that key features of the ecosystem and hydrological regime and their interactions at multiple sites are used as basis for EFA; EFR process includes: (1) identification of management performance criteria by panel of experts for 5 main ecosystem components: fish, trees, macrophytes, invertebrates and geomorphology, (2) application of the criteria for three elements (and associated descriptors) identified as exerting an influence on the ecosystem components (viz. flow regime, hydrograph and physical structure at 3 spatial scales), (3) workshop-based cross-tabulation approach to identify and document generalised responses and/or impacts for each ecosystem components to each specific descriptor (for each element), so as to relate flow regime attributes to ecosystem responses and EFRs; incorporates system hydrological variability and elements of ecosystem functioning; includes stakeholder-panel member workshop for EFR refinement; well defined EFA objectives; some potential for inclusion of other ecosystem components; led to the evolution of other expert-panel approaches; limited use of field data; poor definition of output format for EFR; moderately rapid, flexible and resource-intensive; simpler, less quantitative supporting evidence and less rigorous than Flow Restoration Methodology, BBM and DRIFT; recent applications and limitations reviewed, need for a Best Practice Framework identified.

Appears limited to a single application in Australia in its original form; general approach variously modified for other expert-panel based EFAs

Habitat Analysis Method:
(Walter et al. 1994; Burgess and Vanderbyl 1996; Arthington 1998).

Developed by former Queensland Department of Primary Industries, Water Resources (now DNR), Australia, as part of water allocation and management planning initiative.

Relatively rapid, inexpensive, basin-wide reconnaissance method for determining preliminary EFRs at multiple points in catchment (rather than at a few critical sites); superior to simple hydrological EFMs, but inadequate for comprehensive EFAs; field data limited or absent; bottom-up process of 4 stages using TAP: (1) identification of generic aquatic habitat types existing within the catchment, (2) determination of flow-related ecological requirements of each habitat (as surrogate for EFRs for aquatic biota), using small group of key flow statistics, plus select ‘biological trigger’ flows and floods for maintenance of ecological and geomorphological processes, (3) development of bypass flow strategies to meet EFRs, (4) development of EFR monitoring strategy; EFM represents an extension of expert panel approaches (EPAM, SPAM), with conceptual basis and assumptions adapted from Holistic Approach; little consideration of specific flow needs of individual ecological components; requires standardisation of process, refinement of flow bands linked to habitats and addition of flow events related to needs of biota; represents a simplified version of the Holistic Approach; largely superseded by Benchmarking Methodology.

Precursor of Benchmarking Methodology within WAMP initiatives; several applications within Australia

Benchmarking Methodology:
(Brizga et al. 2001, 2002).

Developed in Queensland, Australia, by local researchers and DNR, to provide a framework for assessing risk of environmental impacts due to WRDs, at basin scale.

Rigorous and comprehensive, scenario-based, top-down approach for application at basin scale; using field and desktop data for multiple river sites; same conceptual basis as BBM and Holistic Approach, EFM has 4 main stages: (1) establishment: formation of multidisciplinary expert panel (TAP) and development of hydrological model for catchment, (2) ecological condition and trend assessment: development of spatial reference framework (multiple river sites within representative and critical river reaches), assessment of ecological condition for suite of ecosystem components (using 3-point rating of degree of change from reference condition and appropriate methods for assessing each component), development of generic models (conceptual, empirical) defining links between flow regime components and ecological processes, selection of key flow indicators and statistics with relevance to these relationships, modelling-based assessment of hydrological impacts, (3) development of risk assessment framework to guide evaluation of potential impacts of future water resource development and management scenarios: benchmark models are developed for all or some key flow indicators showing levels of risk of geomorphological and ecological impacts associated with different degrees of flow regime change, risk levels are defined by association with benchmark sites which have undergone different degrees of flow-related change in condition, link models are used to show how the modelled flow indicators affect ecological condition, (4) evaluation of future WRD scenarios, using risk assessment and link models, ecological implications of scenarios and associated levels of risk readily expressed in graphical form; EFM is particularly suited to data poor situations; potential for use in developing countries and for application to other aquatic ecosystems (e.g. wetlands, estuaries); utilises a wide range of specialist expertise; presents a comprehensive benchmarking process and transparent reporting system; provides several ways of developing risk assessment models, guidance on key criteria for assessing condition and key hydrological and performance indicators; a recent approach built on several preceding EFA initiatives; no explicit consideration of social component, but with scope for inclusion of socio-economic assessments (note that socio-economic issues are evaluated separately by DNR and considered when the final EF recommendations are being decided); requires evaluation of several aspects (e.g. applicability or sensitivity of key flow statistics, degree to which benchmarks from other basins or sites within basins are valid considering differences in river hydrology and biota); recommends a monitoring programme and additional research on important issues, as crucial components of EF implementation; requires documentation of generic procedure for wider application.

Sole holistic EFM for basin-scale assessment and assessing risk of environmental impacts due to WRD; adopted for routine application in Queensland with applications in 15 basins; under consideration for use in Western Australia; only applied in Australia to date

Environmental Flow Management Plan Method (FMP):
(Muller 1997; DWAF 1999).

Developed in South Africa by the Institute for Water Research, for use for intensively regulated river systems.

Simplified bottom-up approach, applicable in highly regulated and managed systems with considerable operational limitations; considered for use within South Africa Reserve determination process only where BBM or equivalent approach cannot be followed; workshop-based, multidisciplinary assessment including ecologists and system operators; 3-step process: (1) definition of operable reaches for study river and site selection, establishment of current operating rules, (2) determination of current ecological status and desired future state, (3) identification of EFRs using similar procedures to BBM; EFM has limited scope for application; structure and procedures for application are not formalised or well documented; poorly established post-workshop scenario phase; no evaluation undertaken; considerably more limited approach than Flow Restoration Methodology.

Limited to 3 applications; only used in South Africa to date; uncertain status within the national Reserve framework

River Babingley (Wissey) Method:
(Petts et al. 1999).

First developed for application in groundwater-dominated rivers, Anglian Region of England.

Bottom-up field and desktop approach; EAFR (the EF regime) defined in 4 stages: (1) ecological assessment of river and specification of an ecological objective comprising specific targets (for river components and biota), (2) determination of 4 general and 2 flood benchmark flows to meet the specified targets, (3) use of flows to construct ‘ecologically acceptable hydrographs’, which may include provision for wet years and drought conditions, (4) assignment of acceptable flow frequencies and durations to the hydrographs and their synthesis into a flow duration curve, the EAFR; EFM uses hydro-ecological models, habitat and hydrological simulation tools to assist in identification of benchmark flows and overall EAFR; allows for flexible examination of alternative EF scenarios; loosely structured approach, with limited explanation of procedures for integration of multidisciplinary input; risk of omission of critical flow events from EAFR; specific to baseflow-dominated rivers and requires further research for use in flashy catchments; requires documentation of generic procedure for wider application.

Relatively limited application to date; general approach appears to have been extended to other EFA studies in the UK

Downstream Response to Imposed Flow Transformations (DRIFT):
(King et al. 2003; Arthington et al. 2003a).

Developed in southern Africa by Southern Waters and Metsi Consultants (with inputs from Australian and southern African researchers) as an interactive scenario-based holistic EFM with explicit socio-economic component.

Rigorous and well-documented top-down, scenario-based process with interactive scenario development; same conceptual basis as BBM and Holistic Approach; appropriate for comprehensive EFAs (1-3 years) based on several sites within representative and critical river reaches; comprised of 4 modules: (1) biophysical module: used to describe present ecosystem condition, to predict how it will change under a range of different flow alterations, uses generic lists of links to flow and relevance for each specialist component, each prediction and the direction and severity of change are recorded in a database, to quantify each flow-related impact, (2) sociological module: used to identify subsistence users at risk from flow alterations and to quantify their links with the river in terms of natural resource use and health profiles, (3) scenario development module: links first 2 modules through querying of database, to extract predicted consequences of altered flows (with potential for presentation at several levels of resolution); this process is used to create flow scenarios (typically 4 or 5), (4) economic module: generates description of costs of mitigation and compensation for each scenario; well developed ability to address socio-economic links to ecosystem; considerable scope for comparative evaluation of alternative modified flow regimes; high potential for application to other aquatic ecosystems; resource intensive but amenable to simplification for more rapid assessments; uses many successful features of other holistic EFMs; exhibits parallels with Benchmarking Methodology; output is more suitable for negotiation of tradeoffs than in BBM or other bottom-up approaches, as implications of not meeting the EFR are readily accessible; links to external public participation process and macro-economic assessment; generic lists provide clear parameters for inclusion in a monitoring programme; applicable to regulated or unregulated rivers and for flow restoration; EFM modules require refinement; approach provides limited consideration of synergistic interactions among different flow events and ecosystem components; limited inclusion of flow indices describing system variability; recommends a monitoring programme and additional research on important issues, as crucial components of EF implementation; requires documentation of generic procedure for wider application.

EFM with most developed capabilities for scenario analysis and explicit consideration of social and economic effects of changing river condition on subsistence users; limited application to date, within southern Africa

Adapted BBM-DRIFT Methodology:
(Steward et al. 2002).

Developed in Zimbabwe by Mott MacDonald Ltd. in collaboration with Zimbabwe National Water Authority (with input from South Africa) through adaptation of key elements of BBM and DRIFT, in response to requirements in new Water Act for EFAs.

Simplified top-down, multidisciplinary team approach, for use in highly resource-limited (including data limited) situations and with direct dependencies by rural people on riverine ecosystems; combines pre-workshop data collection phase of BBM with DRIFT’s scenario-based workshop process; comprises 3 phases: (1) preparation for workshop as per BBM and DRIFT, but excluding certain components (e.g. habitat integrity and geomorphological reach analyses) and with limited field data collection, (2) workshop, with simplified DRIFT process linking the main geomorphological, ecological and social impacts with elements of the flow regime (based on assessments of impact and severity for component-specific generic lists), used to construct a matrix, (3) use of matrix in evaluating development options, where the matrix indicates ecosystem aspects that are especially vulnerable or important to rural livelihoods, socially and ecologically critical elements of the flow regime and EF recommendations for mitigation; EFM incorporates more limited ecological and geomorphological assessments than BBM and DRIFT; limited coverage of key specialist disciplines; no link to system for defining target river condition; limited capability for scenario development; especially appropriate in developing countries context; requires further development and validation; would benefit from inclusion of economic data.

Under early development, single documented application to date

Flow Restoration Methodology (FLOWRESM):
(Arthington et al. 1999; Arthington et al. 2000).

Developed in a study of the Brisbane River, Queensland, Australia, specifically addressing EFRs in river systems exhibiting a long history of flow regulation and requiring flow restoration.

Primarily bottom-up, field and desktop approach appropriate for comprehensive (or intermediate) EFAs; EFM represents hybrid of Holistic Approach and BBM; designed for use in intensively regulated rivers with emphasis on identification of the essential features that need to be built back into the hydrological regime to shift the regulated river system towards the pre-regulation state; EFM uses an 11-step process in 2 stages, in which the following are achieved: (1) review of changes to the river hydrological regime (focusing on unregulated, present day and future demand scenarios, using hydrological simulation model), (2) series of 8 steps within scenario-based workshop, using extensive multidisciplinary specialist input from field work, literature and expert judgement: determination of flow-related environmental effects for low and high flow months, rationale and potential for restoration of various flow components so as to restore ecological components and functions and establishment of EFRS based on identification of critical flow thresholds or flow bands that meet specified ecological or other objectives, (3) develops series of EF scenarios (quantity, timing, duration of flows) and assesses implications of multiple scenarios for system yield, (4) outlines remedial actions not related to flow regulation, alternatives to flow restoration (e.g. physical habitat restoration, fish passage facilities) are evaluated when some elements of pre-regulation flow regime cannot be restored fully for practical or legal reasons, (5) outlines monitoring strategy to assess benefits of EFRs; particular relevance to rivers regulated by large dams, but applicable to any river system regulated by infrastructure or surface and/or groundwater abstraction; includes well-developed hydrological and ecological modelling tools; more rigorous than expert-panel methods; includes flexible top-down process for assessing ecological implications of alternative modified flow regimes and impacts of not restoring particular flows; potential for adoption of full benchmarking process to rank outcomes of not restoring critical flows; some risk of inadvertent omission of critical flow events (common to all holistic approaches); requires documentation of generic procedure for wider application.

Most comprehensive EFM for flow-related river restoration; single application in Australia to date; EFM case study on Brisbane River used as a procedural guide in other recent EF applications (e.g. Ord River study, Western Australia)

Flow Events Method (FEM):
(Stewardson and Cottingham 2002).

Developed by Australian Cooperative Research Centre for Catchment Hydrology to provide state agencies with a standard approach for EFAs.

Top-down method for regulated rivers; considers the maximum change in river hydrology from natural or key ecologically relevant flow events, based on empirical data or expert judgement; considered a method of integrating existing analytical techniques and expert opinion to identify important aspects of the flow regime; EFM comprises 4 steps: (1) identification of ecological processes (hydraulic, geomorphic and ecological) affected by flow variations at range of spatial and temporal scales, (2) characterisation of flow events (e.g. duration, magnitude) using hydraulic and hydrological analyses, (3) description of the sequence of flow events for particular processes, using a frequency analysis to derive event recurrence intervals for a range of event magnitudes, (4) setting of EF targets, by minimising changes in event recurrence intervals from natural or reference or to satisfy some constraint (e.g. maximum percent permissable change in recurrence interval for any given event magnitude); EFM’s singular development appears to be analysis of changes in event recurrence intervals with altered flow regimes; draws greatly on established procedures of other complex EFMs (e.g. BBM, FLOWRESM and DRIFT); may be used to: (1) assess the ecological impact of changes in flow regimes, (2) specify EF management rules and/or targets, (3) optimise flow management rules to maximise ecological benefits within constraints of existing WRD schemes; possibly places undue emphasis on frequency compared with other event characteristic independent of an associated expert panel method, but could be embedded into one as routine procedure.

Recent approach with few applications in Australia to date; often linked to expert-panel approaches

Additional holistic methodologies developed and applied elsewhere include the River Babingley Method (Petts et al. 1999) developed in England and the Adapted BBM-DRIFT methodology developed in Zimbabwe (Steward, Madamombe and Topping 2002).

In applications of holistic methodologies to date, the focus has almost entirely been on river systems, with most effort addressed to the main river channel and its tributaries and it is only relatively recently that specialist methods have been proposed to address the freshwater flow requirements of downstream receiving waterbodies (e.g. floodplains and terminal lakes in large arid-zone and tropical rivers) and estuaries (e.g. Loneragan and Bunn 1999). Further, methodologies to integrate the dynamic interactions of surface and groundwater systems into existing holistic methodologies are at a fairly immature stage of development, with none routinely applied as part of holistic assessments (King et al. 1999).

SHARED STRENGTHS OF HOLISTIC METHODOLOGIES

Most holistic methodologies employ some form of expert panel-based approach in the derivation of the EFRs of rivers, including those that are specifically termed ‘expert panel’ methods in their own right themselves (e.g. Expert Panel Assessment Method and Scientific Panel Assessment Method, see Table 1). In a review of the use and utility of Australian expert panel methods, Cottingham et al. (2002) commented that environmental flow methods using scientific panels have been "an excellent knowledge exchange mechanism", many are "rapid and inexpensive compared to empirical investigations" (but note that the most recent holistic methodologies use empirically derived, as well as other, knowledge sources), have "the flexibility to adapt the most appropriate and up-to-date assessment methods", and can " make use of information ranging from anecdotal to theoretical". Their shortcomings are judged to fall into two categories: those relating to the scope and quality of field assessments and "problems relating to panel discussions and recommendations" (Cottingham et al. 2002).

To offset these shortcomings, Cottingham et al. (2002) suggest that scientific panel methods would be bolstered by the development of a flexible "best practice" approach suitable for wide application and including the following major features:

It is worth noting that a "Best Practice Framework" (see Figure 1) for the conduct of holistic environmental flow assessments is already available in Australia (Arthington et al. 1998) but was not discussed in the review by Cottingham et al. (2002) even though it appears to offer most of the recommended elements of good practice for ‘scientific panel’ methods. (2002). In the following sections of this paper, we show how the more sophisticated and structured holistic methodologies share common features that address the best practices recommended above (points 1-7) and the common and additional features proposed by Arthington et al. (1998). We focus particularly on the BBM, DRIFT and the Benchmarking and Flow Restoration methodologies, as these represent the most recent developments in holistic methodologies familiar to us and were not included in the appraisal of Cottingham et al. (2002).

Figure 1. Best Practice Framework for assessing environmental flows in regulated and unregulated river systems (from Arthington et al. 1998).

1) Protocols for selecting scientific/expert panels

Guidelines for selecting scientific or technical panel members were established as part of the BBM (King et al. 2000) and these formal procedures have been incorporated into all of the more recent holistic methodologies often based upon the well-established protocols of the BBM (King et al. 2000). Each assessment using the BBM, DRIFT, Benchmarking and Flow Restoration methodologies involves one or more scientists in the fields of hydrology (and occasionally, geohydrology), hydraulics, geomorphology, water quality and aquatic ecology (algae and aquatic plants, riparian vegetation, invertebrates, fish, and wildlife and occasionally, estuarine ecology). Each scientist is expected to be familiar with the river system under study or similar types of rivers and/or EFA procedures.

The roles, responsibilities and interactions of panel members during EFA studies and associated workshops are governed by the particular step-by-step procedures built into each methodology. These procedures generally circumvent outright dominance of workshops and discussions by any one member of the team. Each member has equal opportunity to contribute as fully as they wish and it is usually not possible for any one member to dominate the workshops or bias the outcomes of the evaluations of environmental flow evaluations. Furthermore, workshops forming part of the BBM, DRIFT, Benchmarking and Flow Restoration methodologies are structured and facilitated in such a way that there are frequent comparisons of results and EFA evaluations and results among the participating scientists. These comparisons generally reveal any inconsistencies of approach, or vastly different rankings of flow-related impacts in terms of one or other ecosystem component (other than inherent differences) and/or areas of personal bias. If such issues can be identified early in the workshop process, they can usually be resolved before any consistent patterns of bias affect the entire EFA process. Sensitivity analysis can also be used to identify the influence of particular components of the overall outcome of an EFA.

2) Guidelines for developing objectives

The BBM, DRIFT, Benchmarking and Flow Restoration methodologies all address clear working objectives established as part of the study design, and formalized in design and the contracts signed between the client and each scientific or technical panel member. One or more shared, broad river visions (desired future states in the BBM) may be established, or several more common water resource development or flow restoration objectives may be set, and EFAs evaluated to achieve these objectives. The common practice is to evaluate the ecological consequences of several well-defined scenarios of change in flow regime (either flow reductions, or degrees of flow restoration). These scenarios may be defined using various hydrological statistics, plots and/or indices describing important features of the flow regime modified versus the natural (unregulated) flow regime. Hydrological statistics generally related to flow quantity, timing, duration, frequency of floods and low flow spells, rates of change (e.g. hydrograph rise and fall) and other aspects of variability, including the presence/absence of definite patterns of flow seasonality (after Richter et al. 1996, 1997), as well as ill-defined objectives leading to weak EF recommendations are less likely when the scenarios of hydrological change are explicitly and statistically defined, and/or desired ecological endpoints are stated at the outset of the study. To aim simply for ‘improved river health’ or ‘a sustainable river ecosystem’ is too imprecise an objective for sound scientific assessment.

3) Guidelines for field work

Cottingham et al. (2002) noted that many scientific panel assessments on rivers of southern Australia are based only on desk-top methods and best-available information (often very limited or of poor quality), or involve little more than a rapid field assessment and single spatial/temporal "snap-shot" of the river system conducted at sites "assumed to be representative of the river system under consideration". In contrast, the site selection procedures of the BBM, DRIFT, Flow Restoration and Benchmarking methodologies have a sound, well-documented rationale and they all offer an explicit and transparent framework and methods, for evaluating the ecological implications of many alternative flow scenarios. A range of quantitative procedures can be applied within any of these methodologies to relate flow changes to ecological responses (e.g. wetted perimeter analysis, vegetation transect analysis, water and sediment budget analyses, empirical statistical models, multivariate statistical analyses, predictive population models). For example, the fish components of the Flow Restoration Methodology and DRIFT involve a year of field studies designed to enable consideration of a core set of flow-related aspects of fish biology/ecology (see Pusey 1998; Pusey et al. 1998; Kennard, Arthington and Thompson 2000; Pusey, Kennard and Arthington 2000; Rall 1999; Arthington et al. 2003a). The Benchmarking Methodology, in contrast, relies heavily on the interpretation of data from past field studies, the literature and professional judgement rather than new field studies to relate the ecological condition of fish communities to the level and type of flow modification (Brizga et al. 2002).

4) Procedures for rating confidence levels

The level of confidence in the BBM, DRIFT and Benchmarking assessments is rated according to the information sources available and their scientific quality, thus providing the water manager with an explicit means to undertake his/her own assessment of the risks associated with management actions based on limited or low quality information. The rating scheme applied in DRIFT closely resembles that recommended by Downes et al. (2000) and adapted by Cottingham et al. (2002) into "levels of evidence that support environmental flow assessments". In addition to confidence ratings, the application of DRIFT in the Lesotho Highlands Project involved several phases of peer review (see King et al. 2003), which parallel the sequence of reviews proposed by Arthington et al. (1998) in their Best Practice Framework for environmental flow assessments.

5) Estimating social and economic consequences

The DRIFT methodology includes an explicit process for evaluating the social consequences of each flow scenario stemming from earlier, less clearly defined procedures applied within the BBM (King et al. 2002) and thereby a means to estimate the economic costs of flow regulation in terms of changes in fish and other natural resources or services used by local rural communities (King et al. 2003). This represents a significant advance of DRIFT over other holistic methodologies and avoids the charge that "scientific panels have only ‘green’ value-systems and that they are an alternative environmental lobby" (Cottingham et al. 2002). The Flow Restoration Methodology (Arthington et al. 2000) and the Best Practice Framework also incorporate socio-economic considerations, the former by including a process for evaluating the ‘cost’ of many different environmental flow scenarios generated by releasing flows from storage. In that study, costs were represented in terms of water yields foregone from a large storage reservoir if particular environmental flow scenarios were to be implemented (Arthington et al. 1999; 2000).

6) Documentation

The BBM, DRIFT, Flow Restoration and Benchmarking methodologies all produce comprehensive literature reviews and data reports describing the study area and its ecological systems, EFA methods, results and recommendations, thereby providing major reference documents and benchmarks upon which to base the planning and design of any river restoration activities and future assessments or post-implementation monitoring of river condition. The collation of historic information and preparation of a sequence of refereed reports is a fundamental aspect of the Best Practice Framework (Arthington et al. 1998).

7) Monitoring and further research

Cottingham et al. (2002) did not recommend the incorporation of an explicit monitoring phase as part of scientific panel assessments, although they alluded to the principles of adaptive environmental management (Walters 1987). These principles and rigorous monitoring protocols are built into most other holistic methodologies (see Table 1) and the Best Practice Framework (Figure 1). For example, all components of DRIFT include a detailed rationale and protocol for monitoring the geomorphological or ecological outcomes of environmental flow allocations and water management scenarios (King et al. 2003; Metsi Consultants 2000). With regard to the application of DRIFT in Lesotho rivers, the predictions of fish responses to each environmental flow scenario have formed the basis of hypotheses for testing by monitoring and longer-term research (J. Rall, pers. comm. 2003). Benchmarking Methodology reports always include a section describing key knowledge gaps and research priorities for the catchment under study and the Flow Restoration Methodology devotes a chapter to research and monitoring requirements.

In considering the recommendations of Cottingham et al. (2002) as to the desirable elements of best practice in holistic EFAs based on ‘scientific panel’ approaches, we suggest that the most recent holistic methodologies developed and applied in Australian and southern Africa already address the main concerns and limitations raised above, as does the Best Practice Framework (Arthington et al. 1998). Even so, all such methodologies can be enhanced in many ways and in the next section of this paper we discuss opportunities for the further development of this type of approach to EFAs, particularly in relation to the methods and models used to predict the ecological consequences of flow regime change.

Further development of holistic methodologies

King et al. (1999) and Tharme (2003) consider holistic methodologies to be especially appropriate for use in developing countries, due to the need for resource protection at an ecosystem scale and the direct dependence of local people on the goods and services provided by aquatic ecosystems for food and broader livelihood security. Arthington et al. (2003a) place holistic methodologies at the second level of a three-tiered hierarchy of EFA methods, reflecting recognition by several colleagues (Tharme 1996; Dunbar et al. 1998) of the levels of complexity, confidence in outcomes and risk of error at which EFAs are needed and applied. These are Level 1: precautionary hydrological approaches; Level 2: Holistic scientific panel methodologies using all types of data, information and professional judgement in a structured framework, usually applied when time/resource constraints preclude lengthy investigations and predictive model development; and, Level 3: EFA assessments based on detailed studies and predictive flow-ecology models. Tharme (2003) rated holistic methodologies as having moderate to high resource intensity, complexity and resolution and high flexibility (Table 1) and recommended their use when assessing water resource developments, typically of large-scale, involving rivers of high conservation and/or strategic importance and/or with complex user tradeoffs.

In assessing the utility of DRIFT and other holistic methodologies, Arthington et al. (2003a) and King et al. (2003) considered the gravest risk to be that such approaches "may be used routinely and become all that is sought and used, rather than investing in securing new knowledge of river ecology to guide sound decision-making in the future". They caution that "DRIFT and other scientific panel methods should only be used where there is a genuine commitment to implement and monitor the recommended environmental flows, to support knowledge development and to adapt water management strategies when better information about the river’s responses to flow modification becomes available through monitoring and research".

Clearly, holistic methodologies can be enhanced by integrating modelled responses of river ecosystems to flow change, be it regulation or restoration, that is, by moving towards Level 3 of the EFA hierarchy outlined above. At this level of resolution, environmental water requirements would be defined and alternative water resource developments or restoration scenarios evaluated, by means of quantitative predictive models describing relationships between hydrology and the flow-related ecological processes governing biological diversity and river ecosystem integrity (Arthington et al. 2003).

Quantitative models that describe associations between flow and geomorphological or ecological parameters are available for some ecosystem components (see Arthington and Zalucki 1998 and literature cited therein). For example, hydraulic geometry models can be used to provide an indication of the likely net change in channel dimensions resulting from flow regime change (Brizga et al. 2001). Sediment transport models can provide an indication of the likely implications of flow regime change for sediment processes. Wetland and riparian water budget analyses have proved useful in environmental flow studies designed to restore regulated stream ecosystems (e.g. Pettit, Froend and Davies 2001).

It is useful to briefly review existing techniques and models that predict the responses of fish to changes in river flow regime and the extent of their application in EFAs and river flow management in general.

Hydraulic rating and habitat simulation methods and modelling packages (e.g. PHABSIM - part of IFIM) have been discussed above, so we confine this review to some of the more advanced approaches. Over a decade ago, O’Brien (1987) defined the minimum stream flow hydrograph to maintain existing habitat, food supplies and spawning potential of the endangered Colorado River squawfish (Ptychocheilus lucius) in terms of four flow characteristics. To develop this minimum hydrograph, O’Brien (1987) combined the results of a two-year field study, a physical model, laboratory simulation of flows over cobble substrate and a mathematical sediment transport model. Hill, Platts and Beschta (1991) developed a method linking the timing and magnitude of the low and high flow attributes of annual flow hydrographs to instream and out of channel physical habitat availability and suitability for fish. In a more ambitious program of studies, Williamson, Bartholow and Stalnaker (1993) developed a conceptual framework and a suite of interactive mathematical models of salmon production (SALMOD) simulating the dynamics of resident and anadromous freshwater populations. Milhous (2003) applied a time series analysis of predicted changes in spawning, incubation, fry and juvenile habitat of brown trout to model temporal changes in abundance associated with discharge. This approach was also used to model the effect of reservoir construction on riparian dynamics.

The most recent developments in fisheries modelling in relation to river hydrology and flow management are outlined in Arthington et al. (2003b) and Halls and Welcomme (2003). Fisheries models can be broadly categorised as empirical, population dynamics and holistic. Empirical models are statistical representations of variables or relationships of interest, without reference to the underlying processes. They have been used to describe the response of fish yield to one or more explanatory variables including measures of river morphology, such as drainage basin or floodplain area (e.g. Welcomme 1985), morpho-edaphic indices (Bayley 1988; Pusey et al. 2000) and fishing intensity (Welcomme 1985; Bayley 1988). Other models of this type describe the relationship between fish catches and freshwater flows into estuaries (Loneragan and Bunn 1999), an approach now forming part of Australian environmental flow assessments in coastal rivers.

Fish population dynamics models attempt to describe the response of fish populations to exploitation and environmental variation based upon established theories of population regulation and upon recent advances in understanding of floodplain-river fisheries ecology and biology (Welcomme and Hagborg 1977; Halls, Kirkwood and Payne 2001; Halls and Welcomme 2003). They have yet to be incorporated into holistic environmental flow assessments in any routine fashion, although efforts to do so are in progress (P. Dugan, pers. comm. 2003). Nevertheless, recent applications have informed river flow management. For example, Minte-Vera (2003) developed a lagged recruitment, survival and growth model for the migratory curimba Prochilodus lineatus (Valenciennes, 1847) in the high Paraná River Basin (Brazil), with recruitment as a function of flooding and stock size. Results obtained were used to evaluate the risk to the population from various fisheries and dam-operation management decisions.

In the field of inland and floodplain fisheries, the term ‘holistic’ applies to models that address fish production or yield in the broader context of environmental management and therefore integrate a diversity of variables of hydrological, environmental or social nature (e.g. fishing methods and effort). Holistic models can be broadly classified into ecological models (e.g. Ecopath, see www.ecopath.org), multi-agent models (e.g. FIRMA 2000; see http://cormas.cirad.fr/indexeng.htm) and Bayesian networks. Baran, Makin and Baird (2003) used a Bayesian network model to assess impacts of environmental factors, fish migration patterns and land use options on fisheries production in the Mekong River. Bayesian network models are slowly being incorporated into decision support systems for the determination of river flow regimes that will sustain river ecosystems and their fish populations.

Despite these advances in fisheries modelling (see also Arthington et al. 2003b; Halls and Welcomme 2003) and modelling developments for other ecosystem components (beyond the scope of this paper), the range of available quantitative models is generally too narrow, or too limited in transferability across river ecotypes, to provide a comprehensive basis for environmental flow determinations. Therefore, models generally need to be used in conjunction with/or as a component within other knowledge-based methodologies. Furthermore, many of the ecological models remain black-box (empirical) models and the ecological processes they represent are not well understood. Quantitative models of the secondary effects of flow regime change (e.g. impacts of channel contraction for vegetation and in-stream biota) are generally not available.

IMPROVING THE KNOWLEDGE BASE FOR ENVIRONMENTAL FLOW ASSESSMENTS

Although major advances have been achieved in the broad field of river ecology in recent decades, substantial information gaps characterize every fundamental aspect of aquatic biology and the ecological processes sustaining aquatic ecosystems are still poorly understood (e.g. Walker, Sheldon and Puckridge 1995; Winemiller 2003), particularly in large floodplain river systems that are most threatened by water resource development, fishing pressure and catchment disturbance (Tockner and Stanford 2003). The main knowledge gaps and research priorities for riverine fish and fisheries have been reviewed by Arthington et al. (2003b) and for aquatic systems more generally by Dugan et al. (2002).

In the following section, we comment on the value of experimental studies and long-term research to inform river management and environmental flow decision-making in particular.

FLOW MANIPULATION EXPERIMENTS

Experimental manipulation of river flow can provide useful information informing environmental flow assessments and some experimental releases from dams have been made in this context (e.g. Harris and Gherke 1995). For example, King, Cambray and Impson (1998) demonstrated that experimental releases from the Clanwilliam Dam on the Olifants River, western South Africa, resulted in spawning and larval recruitment of the Clanwilliam Yellowfish (Barbus capensis), provided that water temperatures were suitable for spawning activity, egg survival and larval development. Additional examples, focused on the effects of managed floods on the floodplain wetlands of large rivers, are provided in Acreman, Farquharson, McCartney et al. (2000).

ENVIRONMENTAL FLOWS AS ECOLOGICAL EXPERIMENTS

There are few opportunities for experimentation in unregulated river systems and the high cost of water has precluded widespread experimentation in many regulated systems. Infrastructure constraints (e.g. size of outlet valves, number of flood control gates) also limit the scope of flow experimentation that is possible. Nevertheless, many scientists argue that the implementation of environmental flow regimes and river restoration projects should be regarded as opportunities to conduct ecological experiments (Kingsford 2000; Lake 2001; Bunn and Arthington 2002) and have called for rigorous and comprehensive monitoring of the ecological outcomes of environmental flows to guide river flow management in the future. Poff et al. (2003) have outlined how large scale demonstration flow restoration projects in focus catchments that have significant problems due to flow regime modification and realistic opportunities for flow restoration, could inform river science and management. Although there are likely to be significant experimental design issues (few suitable reference systems and limited opportunities for replication), ecologists believe that turning flow restoration projects into experiments in restoration ecology should be part of the research agenda informing river flow management and are long overdue (Kingsford 2000; Lake 2001; Bunn and Arthington 2002).

LONG-TERM ECOLOGICAL RESEARCH

Long-term research in relative undisturbed catchments appears essential to improve our understanding of river ecosystem functioning in relation to hydrological history and flow events such as floods and droughts. From appropriate spatial and time series investigations it may eventually be possible to develop suites of models predicting how rivers will respond to natural flow variations (and climate change) and various types of flow regulation (Kingsford 2000; Bunn and Arthington 2002; Arthington and Pusey 2003). Such research is also needed to strengthen predictions of restoration trajectories after flows are restored to regulated rivers and their floodplains (Petts 1987; Ward et al. 2001; Lake 2001). With further climate change likely, river flow regimes will change in response to altered thermal and rainfall distributions, increasing evaporation rates, more extreme floods and droughts and increasing water stress. Water shortages and increasing competition for the available water will place even greater demands on the scientific community to define (and defend) the flow requirements of rivers and floodplains.

CONCLUSIONS AND RECOMMENDATIONS

This paper has outlined the origins and development of four types of environmental flow methodology recognised by Tharme (1996; 2003), namely hydrological, hydraulic rating, habitat simulation and holistic approaches. The latter category of methods, of which there are now 16 different types, have many features and strengths in common, particular the use of a multi-disciplinary team of scientists and the structured analysis of EFRs, usually in a workshop setting. We have shown that the most recent holistic methodologies - BBM, DRIFT, Benchmarking and Flow Restoration - already address and in some aspects improve upon, the main elements of best practice in holistic EFAs recommended by Cottingham et al. (2002). An existing Best Practice Framework (Arthington et al. 1998) sums up the most desirable elements of holistic EFAs and most of the new generation holistic EFAs conform to this model.

Nevertheless, holistic methodologies could be vastly enhanced by applying a wider range of quantitative techniques to relate flow alterations to ecological responses and by integrating models that facilitate prediction of the responses of river ecosystems to flow change, that is, by moving towards Level 3 of the EFA hierarchy proposed originally by Tharme (1996) and adapted by Arthington et al. (1998, 2003a). At this level of resolution, environmental water requirements would be defined and alternative water resource developments or restoration scenarios evaluated, by means of detailed studies and quantitative predictive models of the relationships between hydrology, biophysical processes and ecosystem functioning (Arthington et al. 2003a).

Several types of modelling facilitate prediction of the responses of fish and fisheries to river hydrology and changes in flow regime, as well as other environmental and social factors. Recent developments in empirical, population and multi-agent modelling are increasingly being applied in river basin studies and projections of the consequences of river flow change. The integration of such modelling tools into existing and enhanced holistic decision support systems is a priority.

Review and synthesis papers contributed to LARS2 (e.g. Arthington et al. 2003b; Junk and Wantzen 2003; Winemiller 2003) have revealed substantial information gaps in every fundamental aspect of aquatic biology and also show that the ecological processes sustaining aquatic ecosystems are still poorly understood, particularly the functioning of large floodplain river systems. Increasing threats to these systems from water resource development, interacting with fishing pressure, catchment disturbance and climate change, highlight the urgency of establishing experimental research and long-term research programs to inform river management and environmental flow decision-making.

We suggest that there is a role for an international research program to advance the scientific basis of environmental flow assessments in rivers intended for future water infrastructure development and in regulated rivers where there are opportunities for partial restoration of the flow regime. The key elements of existing holistic methodologies discussed in this paper could provide the foundations for new and improved decision support systems, featuring bottom-up and top-down environmental flow methodologies that embody predictive models describing the relationships between river hydrology and flow-related geomorphological and ecological responses. Predictive models of biophysical processes already in use in fisheries management, for example, could be incorporated into EFAs and decision support systems. These models should be linked to processes for assessing the social and livelihoods (and ultimately economic) consequences of changes in flow regimes, for people dependent upon rivers for, among other things, clean water supplies, food resources, fibres, recreational opportunities and spiritual values.

Many features of DRIFT in its current, or variously adapted forms and several Australian holistic methodologies, provide suitable platforms and techniques for the further development of enhanced environmental flow decision support systems. The application of these new generation decision support tools within large scale demonstration flow restoration projects in focus catchments could inform river science and management in both the short and longer term.

ACKNOWLEDGEMENTS

Financial support to Angela Arthington from the organisers of LARS 2 is gratefully acknowledged. Angela and Brad Pusey also thank Griffith University for support to attend LARS 2. Comments from Robin Welcomme and an anonymous reviewer on an early draft of this paper are appreciated and thanks are due also to colleagues who assisted with final revisions and to Samantha Capon for drawing Figure 1.

REFERENCES

Acreman M.C., Farquharson F.A.K., McCartney M.P., Sullivan C., Campbell K., Hodgson N., Morton J., Smith D., Birley M., Knott D., Lazenby J., Wingfield R. & Barbier E.B. 2000. Managed flood releases from reservoirs: Issues and guidance. Report to DFID and the World Commission on Dams. Wallingford, UK, Centre for Ecology and Hydrology. 86 pp.

Arthington A.H. 1998. Comparative evaluation of environmental flow assessment techniques: Review of holistic methodologies. LWRRDC Occasional Paper 26/98. Canberra, Land and Water Resources Research and Development Corporation. 46 pp.

Arthington A.H., Brizga S.O. & Kennard M.J. 1998. Comparative Evaluation of Environmental Flow Assessment techniques: Best practice framework. LWRRDC Occasional Paper 25/98. Canberra, Land and Water Resources Research and Development Corporation. 26 pp.

Arthington A.H., Brizga, S.O., Kennard M.J., Mackay S.J., McCosker R.O., Choy S.C., Ruffini J.L. 1999. Development of a flow restoration methodology (FLOWRESM) for determining environmental flow requirements in regulated rivers using the Brisbane River as a case study. In Proceedings of Hydrology 1999, the 24th Hydrology and Water Resources Symposium. Barton, ACT, Australian Institution of Engineers. pp. 449-454.

Arthington A.H., Brizga S.O., Choy S.C., Kennard M.J., Mackay S.J., McCosker R.O., Ruffini J.L. & Zalucki J.M. 2000. Environmental flow requirements of the Brisbane River downstream from Wivenhoe Dam. Brisbane, Australia, South East Queensland Water Corporation and Centre for Catchment and In- Stream Research. Griffith University. 536 pp.

Arthington A.H., King J.M., O’Keefe J.H., Bunn S.E., Day J.A., Pusey B.J., Bluhdorn D.R. & Tharme R. 1992. Development of an holistic approach for assessing environmental flow requirements of riverine ecosystems. In J.J. Pigram & B.P. Hooper eds. Proceedings of an International Seminar and Workshop on Water Allocation for the Environment. Armidale, USA, Centre for Water Policy Research, University of New England. pp. 69-76.

Arthington A.H., Lorenzen K., Pusey B.J., Abell R., Halls A.S., Winemiller K.O., Arrington D.A., Baran E. 2003b. River fisheries: Ecological basis for management and conservation. In R.L. Welcomme & T. Petr, eds. Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Volume 1. Food and Agriculture Organization of the United Nations & Mekong River Commission. FAO Regional Office for Asia and the Pacific, Bangkok. RAP Publication 2004/16. pp. 21-60.

Arthington A.H. & Pusey B.J. 1993. In-stream flow management in Australia: Methods, deficiencies & future directions. Australian Biologist, 6: 52- 60.

Arthington A.H. & Pusey B.J. 2003. Flow restoration and protection in Australian rivers. River Research and Applications, 19: 377-395.

Arthington A.H., Rall J.L., Kennard M.J. & Pusey B.J. 2003b2. Assessment of the environmental flow requirements of fish in Lesotho rivers using the DRIFT methodology. River Research and Applications, 19: 641-666.

Arthington A.H. & Zalucki J.M. eds. 1998. Comparative evaluation of environmental flow assessment techniques: Review of methods. LWRRDC Occasional Paper Series 27/98. Canberra, Land and Water Resources Research and Development Corporation. 141 pp.

Baran E., Makin I. & Baird I.G. 2003. Bayfish: A model of environmental factors driving fish production in the lower Mekong Basin. Abstract submitted to the Second International Symposium on the Management of Large Rivers for Fisheries. Phnom Penh, Mekong River Commission.

Bayley P.B. 1988. Accounting for effort when comparing tropical fisheries in lakes, river-floodplains and lagoons. Limnology & Oceanography, 33: 963-972.

Bovee K.D. 1982. A Guide to stream habitat analysis using the instream flow incremental methodology. Instream Flow Information Paper 12. Fort Collins, US Department of Fisheries and Wildlife Service. 248 pp.

Brizga S.O., Arthington A.H., Choy S., Craigie N.M., Mackay S., Poplawski W., Pusey B.J. & Werren G. 2001. Pioneer valley water resource plan: Proposed environmental flow assessment framework. Brisbane, Australia, Department of Natural Resources and Mines, 2 Volumes.

Brizga S.O., Arthington A.H., Pusey B.J., Kennard M.J., Mackay S.J., Werren G.L., Craigie N.M., Choy S.J. 2002. Benchmarking, a ‘top-down’ methodology for assessing environmental flows in Australian rivers. Proceedings of International Conference on Environmental Flows for Rivers. Cape Town, SA, University of Cape Town (Available on CD)

Bunn S.E. & Arthington A.H. 2002. Basic principles and ecological consequences of altered flow regimes for aquatic biodiversity. Environmental Management, 30: 492-507.

Burgess G.K. & Vanderbyl T.L. 1996. Habitat analysis method for determining environmental flow requirements. In Proceedings of Water and the Environment, the 23rd Hydrology and Water Resources Symposium. Barton, ACT, Australian Institution of Engineers. 203-206.

Cottingham P., Thoms M.C. & Quinn G.P. 2002. Scientific panels and their use in environmental flow assessment in Australia. Australian Journal of Water Resources, 5: 103-111.

Davies P.M., Bunn S.E. & Arthington A.H. 1996. Environmental water requirements for lowland river systems on the Swan Coastal Plain. Unpublished report to the Water and Rivers Commission, Western Australia. Perth, University of Western Australia, Department of Zoology. 114 pp.

Downes B., Barmuta L., Fairweather P., Faith D.P., Keought M.J., Lake P.S., Mapstone B.D. & Quinn G.P. 2002. Monitoring ecological impacts: Concepts and practice in flowing waters. Cambridge, UK, Cambridge University Press. 434 pp.

Dugan P.J., Baran E., Tharme R., Prein M., Ahmed M., Amerasinghe P., Bueno P., Brown C., Dey M., Jayasinghe G., Niasse M., Nieland A., Smakhtin V., Tinh N., Viswanathan K. & Welcomme R.L. 2002. The contribution of aquatic ecosystems and fisheries to food security and livelihoods: A research agenda. In Background papers of the Challenge Programme on Water and Food. CGIAR. 87- 113.

Dunbar M.J., Gustard A., Acreman M. & Elliott C.R.N. 1998. Overseas approaches to setting river flow objectives. R & D Technical Report W6B964. Wallingford, UK, Institute of Hydrology. 80 pp.

DWAF 1999. Resource directed measures for protection of water resources. Volume 2: Integrated manual. Version 1.0. Pretoria, SA, Department of Water Affairs and Forestry, Institute for Water Quality Studies. 45 pp.

Gippel C.J. & Stewardson M.J. 1998. Use of wetted perimeter in defining minimum environmental flows. Regulated Rivers: Research & Management, 14: 53-67.

Groshens T.P. & Orth D.J. 1994. Transferability of habitat suitability criteria for smallmouth bass, Micropterus dolomieu. Rivers, 4: 194-212.

Halls A.S., Kirkwood G.P. & Payne A.I. 2001. A dynamic pool model for floodplain river fisheries. Ecohydrology and Hydrobiology, 1: 323-339.

Halls A.S. & Welcomme R.L. 2003. Dynamics of river fish populations in response to hydrological conditions: A simulation study. Abstract submitted to the Second International Symposium on the Management of Large Rivers for Fisheries. Phnom Penh, Mekong River Commission.

Hill M.T., Platts S. & Beschta R.L. 1991. Ecological and geomorphological concepts for instream and out-of-channel flow requirements. Rivers, 2: 198-210.

Jowett I.G. 1997. Instream flow methods: A comparison of approaches. Regulated Rivers: Research & Management, 13: 115-127.

Junk W.J. & Wantzen K.M. 2003. The Flood Pulse Concept: New aspects, approaches and applications, an update. In R.L. Welcomme & T. Petr, eds. Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Volume 1. Food and Agriculture Organization of the United Nations & Mekong River Commission. FAO Regional Office for Asia and the Pacific, Bangkok. RAP Publication 2004/16. pp. 117- 140.

Kennard M.J., Arthington A.H. & Thompson C. 2000. Flow requirements of freshwater fish. In Environmental flow requirements of the Brisbane River downstream from Wivenhoe Dam, Brisban. South East Queensland Water Corporation and Centre for Catchment and In- Stream Research.

King J.M., Brown C.A. & Sabet H. 2003. A scenariobased holistic approach to environmental flow assessments for rivers. River Research and Applications, 19: 619-640.

King J.M., Cambray J.A. & Dean Impson N. 1998. Linked effects of dam-released floods and water temperature on spawning of the Clanwilliam yellowfish Barbus capensis. Hydrobiologia, 384: 245-265.

King J.M. & Louw D. 1998. Instream flow assessments for regulated rivers in South Africa using the building block methodology. Aquatic Ecosystem Health and Restoration, 1: 109- 124.

King J.M. & Tharme R.E. 1994. Assessment of the Instream Flow Incremental Methodology (IFIM) and initial development of alternative instream flow methodologies for South Africa. Water Research Commission, Report No. 295/1/94. Pretoria, SA. 590 pp.

King J.M., Tharme R.E. & Brown C.A. 1999. Definition and implementation of instream flows. Thematic Report for the World Commission on Dams. Cape Town, SA, Southern Waters Ecological Research and Consulting. 63 pp.

King J.M., Tharme R.E. & de Villiers M.S. eds. 2002. Environmental flow assessments for rivers: Manual for the building block methodology. Water Research Commission Technology Transfer Report No. TT131/00. Pretoria, SA, Water Research Commission. 340 pp.

Kingsford R.T. 2000. Ecological impacts of dams, water diversions and river management on floodplain wetlands in Australia. Austral Ecology, 25: 109-127.

Lake P.S. 2001. On the maturing of restoration: Linking ecological research and restoration. Ecological Management and Restoration, 2: 110-115.

Loneragan N.R. & Bunn S.E. 1999. River flows and estuarine ecosystems: Implications for coastal fisheries from a review and a case study of the Logan River, Southeast Queensland. Australian Journal of Ecology, 24: 431-40.

Metsi Consultants. 2000. Consulting services for the establishment and monitoring of the instream flow requirements for river courses downstream of LHWP dams. Fish. Report No. 648- F-18. Lesotho, Lesotho Highlands Development Authority Contract 648. 85 pp. plus Appendices.

Milhous R.T., Updike M.A. & Schneider D.M. 1989. Physical habitat simulation system reference manual, version 2. Instream Flow Information Paper 26. U.S.D.I. Fish Wildlife Service Biological Report 8916.

Minte-Vera C.V. 2003. Risk assessment of a floodplain migratory fish subject to fisheries and damoperation procedures. Abstract submitted to the Second International Symposium on the Management of Large Rivers for Fisheries. Phnom Penh, Mekong River Commission.

Naiman R.J., Bunn S.E., Nilsson C., Petts G.E., Pinay G. & Thompson L.C. 2002. Legitimizing fluvial ecosystems as users of water. Environmental Management, 30: 455-467.

O’Brien J.S. 1987. A case study of minimum streamflow for fishery habitat in the Yampa River. In C.R. Thorne, J.C. Bathurst & R.D. Hey RD eds. Sediment Transport in gravel-bed rivers. Chichester, UK, John Wiley & Sons. pp. 921- 946.

O’Keeffe J.H., Hughes D.A. & Tharme R.E. 2001. Linking ecological responses to altered flows for use in environmental flow assessments: The flow stress-response method. Verh. Int. Ver. Limnol. (Available on CD)

O’Keeffe J.H. & Hughes D.A. 2002. The flow stressresponse method for analysing flow modifications: Applications and developments. Proceedings of International Conference on Environmental Flows for Rivers. Cape Town, SA, University of Cape Town. (Available on CD)

Olden J.D. & Poff N.L. 2003. Redundancy and the choice of hydrological indices for characterising streamflow regimes. River Research and Applications, 19: 101-121.

Pettit N.E., Froend R.H. & Davies P.M. 2001. Identifying the natural flow regime and the relationship with riparian vegetation for two contrasting Western Australian rivers. Regulated Rivers: Research & Management, 17: 201-215.

Petts G.E. 1987. Time-scales for ecological changes in regulated rivers. In J.F. Craig & J.B. Kemper eds. Regulated streams Advances in ecology. New York, Plenum Press. pp. 257-266.

Petts G.E., Bickerton M.A., Crawford C., Lerner D.N. & Evans D. 1999. Flow management to sustain groundwater-dominated stream ecosystems. Hydrological Processes. 13: 497-513.

Poff N.L., Allan J.D., Bain M.B., Karr J.R., Prestegaard K.L., Richter B.D., Sparks R.E. & Stromberg J.C. 1997. The natural flow regime, a paradigm for river conservation and restoration. BioScience, 47: 769-784.

Poff N. LeRoy, Allan J.D., Palmer M.A., Hart D.D., Richter B.D., Arthington A.H., Rogers K.H., Meyer J.L. & Stanford J.A. 2003. River flows and water wars: Emerging science for environmental decision-making. Frontiers in Ecology and the Environment, 1: 298-306.

Puckridge J.T., Sheldon F., Walker K.F. & Boulton A.J. 1998. Flow variability and the ecology of large rivers. Marine and Freshwater Researc,h 49: 55-72.

Pusey B.J. 1998. Methods addressing the flow requirements of fish. In A.H. Arthington & J.M. Zalucki eds. Comparative evaluation of environmental flow assessment techniques: Review of methods. LWRRDC Occasional Paper Series 27/98. Canberra, Land and Water Resources R & D Corporation. pp. 66-105.

Pusey B.J., Kennard M.J. & Arthington A.H. 2000. Discharge variability and the development of predictive models relating stream fish assemblage structure to habitat in north-eastern Australia. Ecology of Freshwater Fishes, 9: 30-50.

Pusey B.J., Kennard M.J., Arthur, J.M. & Arthington A.H. 1998. Quantitative sampling of stream fish assemblages: Single versus multiple pass electroshocking. Australian Journal of Ecology, 23: 365-374.

Rall J.L. 1999. Development of a conservation program on Pseudobarbus quathlambae in the catchment area of phase 1B of the Lesotho Highlands Water Project. Rand Afrikaans University. (Doctoral dissertation)

Richter B.D., Baumgartner J.V., Powell J. & Braun D.P. 1996. A method for assessing hydrologic alteration within ecosystems. Conservation Biology, 10: 1-12.

Richter B.D., Baumgartner J.V., Wigington R. & Braun D.P. 1997. How much water does a river need? Freshwater Biology, 37: 231-249.

Rosenberg D.M., McCully P. & Pringle C.M. 2000. Global-scale environmental effects of hydrological alterations: Introduction. BioScience, 50: 746-751.

Sparks R.E. 1992. Risks of altering the hydrologic regime of large rivers. In J. Cairns, B.R. Niederlehner & D.R. Orvos eds. Predicting Ecosystem Risk. Vol XX. Princeton, NJ, Princeton Scientific Publishing Co. pp. 119-152.

Sparks R.E. 1995. Need for ecosystem management of large rivers and floodplains. BioScience, 45: 168-182.

Stalnaker C., Lamb B.L., Henriksen J., Bovee K. & Bartholow J. 1994. The instream incremental methodology. A primer for IFIM. Biological Report, 29. Washington, DC, U.S. Department of the Interior, National Biological Service.

Steward H.J., Madamombe E.K. & Topping C.C. 2002. Adapting environmental flow methodologies for Zimbabwe. Proceedings of International Conference on Environmental Flows for Rivers. Cape Town, SA, University of Cape Town (Available on CD)

Stewardson M.J. & Cottingham P. 2002. A demonstration of the flow events method: Environmental flow requirements of the Broken River. Australian Journal of Water Resources, 5: 33- 47.

Stewardson M.J. & Gippel C. 1997. In-stream environmental flow design: A review. Victoria, Australia, Cooperative Research Centre for Catchment Hydrology, Department of Civil and Environmental Engineering, University of Melbourne. 99 pp.

Swales S. & Harris J.H. 1995. The Expert Panel Assessment Method EPAM: a new tool for determining environmental flows in regulated rivers. In D.M. Harper & A.J.D. Ferguson eds., The Ecological Basis for River Management. Chichester, UK, John Wile & Sons. pp. 125- 134.

Tharme R.E. 1996. Review of international methodologies for the quantification of the instream flow requirements of rivers. Water law review final report for policy development, for the Department of Water Affairs and Forestry. Pretoria, SA, Freshwater Research Unit, University of Cape Town. 116 pp.

Tharme R.E. 2003. A global perspective on environmental flow assessment: Emerging trends in the development and application of environmental flow methodologies for rivers. River Research and Applications, 19: 397-442.

Tharme R.E. & King J.M. 1998. Development of the building block methodology for instream flow assessments and supporting research on the effects of different magnitude flows on riverine ecosystems. Pretoria, SA, Water Research Commission Report No. 576/1/98. 452 pp.

Thoms M.C., Sheldon F., Roberts J., Harris J., Hillman T.J. 1996. Scientific panel assessment of environmental flows for the Barwon-Darling River. Sydney, Australia, New South Wales Department of Land and Water Conservation. 161 pp.

Tockner K. & Stanford J.A. 2002. Riverine flood plains: Present state and future trends. Environmental Conservation, 29: 308-330.

Walker K.F., Sheldon F. & Puckridge J.T. 1995. An ecological perspective on dryland rivers. Regulated Rivers: Research & Management, 11: 85-104.

Walter A.C., Burgess G.K. & Johnston P.J. 1994. Assessment of a process for determining environmental flows. In Environmental Flows Seminar Proceedings. Artarmon, Victoria: AWWA Inc., pp. 195-201.

Walters C. 1997. Challenges in adaptive management of riparian and coastal systems. Conservation Ecology. (Available at http://www.consecol. org/vol1/iss2/art1://, 24 October 2003)

Ward J.V., Tockner U., Uehlinger U. & Malard F. 2001. Understanding natural patterns and processes in river corridors as the basis for effective river restoration. Regulated River: Research and Management, 117: 311-323.

Welcomme R.L. 1985. River fisheries. Fisheries Technical Paper. Rome, Food and Agriculture Organization of the United Nations. pp. 262 - 330.

Welcomme R.L. & Hagborg D. 1977. Towards a model of a floodplain fish population and its fishery. Environmental Biology of Fishes, 2: 7-24

Welcomme R.L. & Halls A.S. 2004. Dependence of tropical river fisheries on river flow. In R.L. Welcomme & T. Petr, eds. Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Volume 2. Food and Agriculture Organization of the United Nations & Mekong River Commission. FAO Regional Office for Asia and the Pacific, Bangkok. RAP Publication 2004/17. pp. 267-284.

Williamson S.C., Bartholow J.M. & Stalnaker C.B. 1993. Conceptual model for quantifying pre-smolt production from flow-dependent physical habitat and water temperature. Regulated Rivers: Research & Management, 8: 15-28.

Winemiller K.O. 2004. Floodplain river food webs: generalizations and implications for fisheries management. In R.L. Welcomme & T. Petr, eds. Proceedings of the Second International Symposium on the Management of Large Rivers for Fisheries Volume 2. Food and Agriculture Organization of the United Nations & Mekong River Commission. FAO Regional Office for Asia and the Pacific, Bangkok. RAP Publication 2004/17. pp. 285- 310.


Previous Page Top of Page Next Page