C S A S
Canadian Science Advisory Secretariat
S C C S
Secrétariat canadien de consultation scientifique
This series documents the scientific basis for the
evaluation of aquatic resources and ecosystems in
Canada. As such, it addresses the issues of the
day in the time frames required and the
documents it contains are not intended as
definitive statements on the subjects addressed
but rather as progress reports on ongoing
La présente série documente les fondements
scientifiques des évaluations des ressources et
des écosystèmes aquatiques du Canada. Elle
traite des problèmes courants selon les
échéanciers dictés. Les documents qu‘elle
contient ne doivent pas être considérés comme
des énoncés définitifs sur les sujets traités, mais
plutôt comme des rapports d‘étape sur les études
Research documents are produced in the official
language in which they are provided to the
This document is available on the Internet at:
Les documents de recherche sont publiés dans
la langue officielle utilisée dans le manuscrit
envoyé au Secrétariat.
Ce document est disponible sur l‘Internet à:
ISSN 1499-3848 (Printed / Imprimé)
ISSN 1919-5044 (Online / En ligne)
© Her Majesty the Queen in Right of Canada, 2013
© Sa Majesté la Reine du Chef du Canada, 2013
Research Document 2012/039
Document de recherche 2012/039
National Capital Region
Région de la capitale nationale
Review of approaches and methods to
assess Environmental Flows across
Canada and internationally
Examen des approches et des méthodes
d’évaluation des débits
environnementaux au Canada et à
Linnansaari, T.1, Monk, W.A.2, Baird, D.J.2 and Curry, R.A.1
1 Canadian Rivers Institute, University of New Brunswick, Department of Biology, P.O. Box 4400,
Fredericton, New Brunswick, E3B 5A3
2 Environment Canada, Canadian Rivers Institute, University of New Brunswick, Department of
Biology, P.O. Box 4400, Fredericton, New Brunswick, E3B 5A3
TABLE OF CONTENTS
ABSTRACT .................................................................................................................................................. iv
RÉSUMÉ ...................................................................................................................................................... vi
1. INTRODUCTION ....................................................................................................................................... 1
1.1 NEEDS FOR ENVIRONMENTAL FLOW FRAMEWORK IN CANADA ........................................ 1
1.2 PURPOSE AND SCOPE OF REVIEW ......................................................................................... 2
2. CONFLICTING TERMINOLOGY OF "ENVIRONMENTAL FLOWS" ....................................................... 3
Instream Flow Need / Requirement ............................................................................................ 3
Environmental Flow ..................................................................................................................... 3
Ecological Flow: .......................................................................................................................... 4
Base Flow: ................................................................................................................................... 4
3. REVIEW OF ENVIRONMENTAL FLOW ASSESSMENT METHODOLOGY ........................................... 4
3.1 HYDROLOGICAL METHODS ...................................................................................................... 5
3.1.1 The Tennant method and its derivatives ............................................................................ 5
3.1.2 Flow Duration (Exceedence) curves, and statistical low-flow frequency methods ............. 6
3.1.3 Indicators of Hydrologic Alteration (IHA) and the Range of Variability Approach (RVA) .... 7
3.1.4. Percentage of Flow (POF) methods, Sustainability Boundary Approach (SBA) and
Presumptive Standards ..................................................................................................... 8
3.1.5 Strengths, weaknesses and data requirements of hydrological methods ........................ 10
3.2 HYDRAULIC RATING METHODS .............................................................................................. 11
3.2.1 General description and common methods ..................................................................... 11
3.2.2. Strengths, weaknesses and data requirements of hydraulic methods ............................ 12
3.3 HABITAT SIMULATION METHODS ........................................................................................... 13
3.3.1 Habitat simulation methods at the microhabitat scale ...................................................... 13
3.3.2 Strengths, weaknesses and data requirements of habitat simulation methods ............... 18
3.3.3 Generalized (statistical) habitat models ........................................................................... 19
3.3.4. Mesohabitat models ........................................................................................................ 20
3.3.5 Bioenergetic models ......................................................................................................... 23
3. 4 HOLISTIC METHODS AND OTHER ANALYTICAL FRAMEWORKS FOR DEVELOPING
ENVIRONMENTAL FLOW STANDARDS ................................................................................ 23
3.4.1 Building Block Methodology (BBM) .................................................................................. 24
3.4.2 Downstream Response to Imposed Flow Transformation (DRIFT) ................................. 25
3.4.3 Benchmarking and the derived frameworks ..................................................................... 27
3.4.4 The Ecological Limits of Hydrologic Alteration (ELOHA) Approach ................................. 28
3.4.5 Strengths, weaknesses and data requirements of holistic methods ................................ 33
4. SUMMARY OF ENVIRONMENTAL FLOW GUIDELINES CURRENTLY USED IN CANADA AND
INTERNATIONALLY .............................................................................................................................. 34
4.1 CURRENT ENVIRONMENTAL FLOW GUIDELINES IN CANADA ........................................... 34
4.1.1 British Columbia ............................................................................................................... 35
4.1.2 Alberta .............................................................................................................................. 35
4.1.3 Saskatchewan and Manitoba ........................................................................................... 36
4.1.4 Ontario .............................................................................................................................. 36
4.1.5 Quebec ............................................................................................................................. 36
4.1.6 Atlantic Provinces (New Brunswick, Nova Scotia, Prince Edward Island and
Newfoundland and Labrador).......................................................................................... 37
4.1.7. Eastern and Western Arctic ............................................................................................ 37
4.2 ENVIRONMENTAL FLOW GUIDELINES IN OTHER SELECT COUNTRIES ........................... 37
4.2.1 USA .................................................................................................................................. 37
4.2.2 European Union ............................................................................................................... 38
4.2.3 Australia, South-Africa and New-Zealand ........................................................................ 38
5. ACKNOWLEDGEMENTS ....................................................................................................................... 39
6. REFERENCES ........................................................................................................................................ 39
TABLES ....................................................................................................................................................... 50
FIGURES .................................................................................................................................................... 58
APPENDIX A: List of Abbreviations ............................................................................................................ 72
APPENDIX B. The current method of establishing minimum or environmental flows in some European
Correct citation for this publication:
Linnansaari, T., Monk, W.A., Baird, D.J. and Curry, R.A. 2013. Review of approaches and
methods to assess Environmental Flows across Canada and internationally. DFO Can.
Sci. Advis. Sec. Res. Doc. 2012/039. vii + 75 p.
Freshwater resources are under increasing threat from anthropogenic activities, both in terms
of consumptive and non-consumptive use. The increasing societal demands for water have led
to substantial flow alterations in rivers in Canada. Flow alteration can be directly linked to
impacts on the physical and chemical attributes and processes of rivers and subsequent
ecological changes. In addition to increasing demand of water, the ecological, social and
cultural values of rivers are increasingly being recognized. With so many competing needs for
water, there is an urgent need to develop sustainable environmental flow management
guidelines to manage the risk associated with alterations to the flow regime in Canadian rivers.
A national environmental flow framework would also support the habitat protection provisions of
the Fisheries Act (S.35).
This document is intended to serve as an input for a scientific review process at a national level
and to provide background information on i) conflicting terminology used in the environmental
flows literature, ii) environmental flow assessment methodological approaches and iii) current
status of environmental flow guidelines used in different jurisdictions in Canada. The
information contained in this document was critically discussed by an assembled group of
experts who will provide further scientific advice on environmental flow regimes for the
Canadian context in a separate document.
The terminology used in environmental flow assessment literature is variable; "instream flow
needs" and "environmental flows" seem to be the most widely used and inclusive terminology.
The endorsement of these terms for wider use in Canadian context needs to be further verified
by an assembled group of experts.
The techniques used in environmental flow regime assessment were classified into four general
categories (hydrological, hydraulic rating, habitat simulation and holistic methodologies and
frameworks) and the benefits and weaknesses are reviewed. The four methodological
categories differ drastically in the scope and implementation costs and therefore, are suited for
different level of assessment of environmental flow regimes. Most assessment methods are not
based on tested relationships between the extent of flow regime alteration and ecological
response. A recent trend seems to suggest that there is an increasing recognition that any
environmental flow method used alone will not be sufficient for determining environmental flows
in all situations; the holistic methods and frameworks (wherein a combination of other methods
are used) are increasingly common especially in large scale projects.
The examination of the current methodologies used in different jurisdictions in Canada revealed
that many provinces do not have an established guideline to be used for determining
environmental flow regimes. Some provinces have established guidelines to be used in
uncontroversial situations (i.e. cases that are believed to lead to no harmful alteration,
disruption or disturbance to fish habitat, "HADD"). None of the jurisdictions appeared to have an
established environmental flow framework that is used in larger-scale projects (i.e. "potential or
incurring HADD") but evaluation is carried out on a case-by-case basis.
Various options for establishing a national environmental flow regime framework are available
but at least two suggestions are to be examined more closely by an assembled group of
experts. The first option is related to incorporating a framework similar to the process
characterized by the Ecological Limits of Hydrologic Alteration (ELOHA; Poff et al. 2010) to at
least some degree in Canada. The second option is to establish a two-tiered framework
wherein the more general Level 1 ("no HADD") assessment would be based on some
combination of hydrologically-based guidelines to protect the natural variability in river flow
regimes with some cut-off values to terminate water withdrawal during the lower flows. The
more specific Level 2 ("potential HADD") assessment would be based on a holistic approach for
which a detailed protocol would need to be drafted to ensure the ecological integrity,
comparability and transparency across Canada. Regardless of the type of framework to be
established, it is fundamental that the established environmental flow standards are preceded
and followed by a controlled monitoring program and the possibility to refine the environmental
flow regime standards by adaptive management in an iterative process.
The need to establish a system to categorize Canadian rivers (or their segments) into
ecological management classes needs to be further discussed by an assembled group of
experts. The potential benefit of such classification would be the possibility to design different
environmental flow standards based on the ecological or societal "value" of various river
sections. Finally, a time-frame must be determined to ensure that a national environmental flow
framework will be established in an expedited manner.
Les ressources d‘eau douce sont de plus en plus menacées par les activités anthropiques, soit
pour des fins de consommation ou non. L‘accroissement de la demande en eau des sociétés a
entraîné la modification importante du débit des rivières au Canada. Cette modification peut
être directement liée aux incidences sur les processus et les caractéristiques physiques et
chimiques des rivières ainsi qu‘aux changements écologiques subséquents. En plus de la
demande accrue en eau, les valeurs écologiques, sociales et culturelles des rivières sont de
plus en plus reconnues. Comme il y a tant de besoins concurrentiels en eau, il est urgent
d‘élaborer des lignes directrices sur la gestion des débits environnementaux durables afin de
gérer le risque lié à la modification du débit des rivières canadiennes. Un cadre national sur les
débits environnementaux appuierait également les dispositions relatives à la protection de
l‘habitat de la Loi sur les pêches (article 35).
Le présent document a pour but de documenter un processus d‘examen scientifique à l‘échelle
nationale et de fournir des renseignements généraux sur i) la terminologie contradictoire utilisée
dans les textes sur les débits environnementaux, ii) les approches méthodologiques utilisées
pour évaluer les débits environnementaux et iii) l‘état actuel des lignes directrices sur les débits
environnementaux utilisées dans les différentes provinces du Canada. Les renseignements
contenus dans ce document ont été discutés par un groupe d‘experts rassemblés. Ces experts
vont ensuite fournir des avis scientifiques sur les débits environnementaux qui s‘appliquent au
contexte canadien dans un document distinct.
La terminologie utilisée dans les textes sur l‘évaluation des débits environnementaux varie. Les
termes « norme de débit minimal » et « débits environnementaux » semblent les plus utilisés et
généraux. Avant qu‘une utilisation plus répandue de ces termes dans le contexte canadien soit
recommandée, un groupe d‘experts devra en réaliser un examen approfondi.
Les techniques d‘évaluation des débits environnementaux utilisées ont été divisées en
quatre grandes catégories (hydrologique, l‘évaluation hydraulique, la simulation de l‘habitat et
les méthodes et les cadres holistiques); on examine actuellement les avantages et les
inconvénients de ces techniques. La portée et les coûts de mise en œuvre des techniques
diffèrent nettement d‘une catégorie à l‘autre. Par conséquent, ces techniques permettent
d‘évaluer les débits environnementaux à différents niveaux. La majorité des méthodes
d‘évaluation ne sont pas fondées sur les relations éprouvées entre la portée de la modification
des débits et la réaction écologique. Selon une tendance récente, il semblerait que l‘on
reconnaisse de plus en plus le fait que toute méthode utilisée seule ne permettra pas de
déterminer les débits environnementaux dans tous les cas; les méthodes et les cadres
holistiques (en combinaison avec d‘autres méthodes) sont de plus en plus utilisés, plus
particulièrement dans le contexte de projets de grande envergure.
L‘examen des méthodes actuellement utilisées dans les différentes provinces du Canada révèle
que plusieurs provinces ne possèdent aucune ligne directrice relative à la détermination des
débits environnementaux. Certaines provinces ont établi des lignes directrices pour les
situations qui ne prêtent pas à controverse (c.-à-d. les cas qui ne devraient pas entraîner la
détérioration, la destruction ou la perturbation de l‘habitat du poisson). Il semble qu‘aucun cadre
sur les débits environnementaux n‘ait été établi dans les provinces pour les projets de grande
envergure (c.-à-d. les cas pouvant entraîner la détérioration, la destruction ou la perturbation de
l‘habitat), mais une évaluation est effectuée au cas par cas.
Il existe plusieurs options pour établir un cadre national sur les débits environnementaux, mais
au moins deux propositions doivent faire l‘objet d‘un examen approfondi par un groupe
d‘experts. La première option vise l‘intégration, au Canada, d‘un cadre semblable au processus
décrit dans le livre Ecological Limits of Hydrologic Alteration (Poff et coll. 2010), dans une
certaine mesure. La deuxième option consiste à établir un cadre à deux volets. L‘évaluation
générale de niveau 1 (aucune détérioration, destruction ou perturbation de l‘habitat) serait
fondée sur une combinaison de lignes directrices axées sur l‘hydrologie afin de protéger la
variabilité naturelle du débit des rivières et comporterait des valeurs limites imposant l‘arrêt de
l‘extraction d‘eau lorsque le débit est faible. L‘évaluation précise de niveau 2 (détérioration,
destruction ou perturbation potentielle de l‘habitat) serait fondée sur une approche holistique
pour laquelle un protocole détaillé devrait être rédigé afin de garantir l‘intégrité écologique, la
comparabilité et la transparence dans l‘ensemble du Canada. Peu importe le type de cadre qui
doit être établi, il est essentiel que l‘établissement des normes sur les débits environnementaux
soit précédé et suivi par un programme de surveillance contrôlée et la possibilité de
perfectionner ces normes en fonction d‘une gestion adaptative dans le cadre d‘un processus
Un groupe d‘experts doit discuter en profondeur de la nécessité de mettre en place un système
permettant de classer les rivières canadiennes (ou leurs parties) dans différentes catégories de
gestion écologique. Cette classification permettrait d‘élaborer différentes normes sur les débits
environnementaux en fonction de la « valeur » écologique ou sociale de diverses parties d‘une
rivière. Enfin, un échéancier doit être établi afin de garantir que le cadre national sur les débits
environnementaux sera élaboré rapidement.
1.1 NEEDS FOR ENVIRONMENTAL FLOW FRAMEWORK IN CANADA
Freshwater resources are under increasing threat from anthropogenic activities, both in terms
of consumptive and non-consumptive use (Poff et al. 2010; Richter et al. 2011). The increasing
societal demands for water have led to substantial flow (i.e. discharge) alterations in rivers both
in Canada and internationally. Flow alteration can be directly linked to impacts on the physical
attributes of rivers and the resulting ecological changes (Figure 1, Clarke et al. 2008; Poff and
Zimmerman 2010). Furthermore, the risk of adverse changes will increase with increasing
magnitude of flow alteration (Poff and Zimmerman 2010). Moreover, the effects of extreme daily
fluctuations such as hydropeaking operations have also been shown to lead to potentially
deleterious effects on numerous biotic and physical components of riverine ecosystems (Bain
Parallel to increasing demands to consume water, the ecological, social and cultural values of
rivers are increasingly being recognized, together with the ecosystem services provided by
"environmental flows" (Richter 2010; see terminology in Section 2 for detailed definitions).
According to the Brisbane Declaration (2007, p. 1), "Freshwater ecosystems are the foundation
of our social, cultural, and economic well-being".
With so many competing needs for water, the question is "How much water does a river need?"
(Richter et al.1997). Moreover, in order to ensure the conservation and protection of fish and
other valued ecosystem components, it has been recognized that the riverine ecosystem is not
only dependent on available water quantity but also on the natural dynamic character of the flow
regime (Poff et al. 1997). The Instream Flow Council (Annear et al. 2004) also stress the need
to maintain the integrity of inter- and intra-annual flow patterns that mimic the natural
hydrograph. Thus, there is an urgent need to develop sustainable, environmental flow
management guidelines to regulate the alterations to the flow regime.
In Canada, the consideration of environmental flow requirements is indirectly mandated by
federal legislation. The federal government, acting through Fisheries and Oceans Canada
(DFO), administers the Fisheries Act. Section 32 of the Fisheries Act prohibits "destruction of
fish by means other than fishing", and Section 35 prohibits "harmful alteration, disruption or
destruction of fish habitat" (i.e. HADD). Despite this legislation, there are currently no federal
guidelines on how to determine environmental flows in Canada, and current recommendations
have been decided ad hoc at regional and provincial levels (Peters et al. 2012). The fact that
there is no existing national framework to set environmental flow standards has led to a
situation where fisheries resources, fish habitat and the supporting freshwater ecosystems may
not be consistently protected across Canada.
With increasing water demand, and potentially changing background levels in water availability
(as predicted from current consensus on the long-term effects of global climate change; IPCC
2007), there is an urgent need to establish such an environmental flows framework in Canada.
This is a challenge, given the large geographic area to be covered which encompasses a
substantial range of hydrological, geomorphological and ecological variety across the country.
Moreover, although many different environmental flow methods exist (over 200; Tharme 2003),
determining an effective approach for the application of environmental flow thresholds across
Canada has proved elusive.
1.2 PURPOSE AND SCOPE OF REVIEW
The purpose of this report is to:
1) Distinguish and comment on the conflicting terminology used in the environmental
2) Provide a summary and comparative review of current environmental flow
assessment methodological approaches; and
3) Collate information on the current status of environmental flow guidelines used in
different jurisdictions in Canada and internationally.
This report served as a draft discussion document for an assembly of national and
international experts in the area of environmental flows science in order to provide
formal scientific advice for the Canadian context (see accompanying Science Advisory
Report published by DFO’s Canadian Science Advisory Secretariat).
It must be noted that the intent of this report is not to review questions related to the
implementation of any potential environmental flow framework or to consider any aspect related
to federal or provincial legislation, or legal responsibilities in various jurisdictions as may be
related to environmental flows. These are critically important elements of effective
environmental flow regime management that will be addressed in other venues.
Due to the wealth of literature concerning environmental flow assessment methods, the
information collated in this report has largely relied on numerous existing review articles.
Greater emphasis was put on the more recent articles describing different approaches. While
not discounting the value of older literature, it was believed that i) the information described in
the older literature would be contained within the numerous review articles on the subject and ii)
the more recent articles are building upon the former knowledge and have potentially
addressed any deficiencies that the older methods may have included. Here we aim to provide
a concise description of general groupings of environmental flow methods. In doing so, we
identify and briefly describe the more commonly-used or emerging methods and frameworks.
As per the status quo of environmental flow guidelines in Canada and the trends internationally,
we limited the scope of this report to considerations that could be collated with "reasonable
effort". This means that the report does not include a listing of every method or case study used
in different jurisdictions, rather, the intent is to provide a general overview of the similarities,
differences and trends related to environmental flow considerations in various jurisdictions.
It is also recognized that defining environmental flow requirements could be very different in
perennial (a stream which flows continuously all year) and intermittent (a stream that flows only
for a part of the year; also seasonal or ephemeral stream) streams and rivers. Environmental
flow frameworks are typically designed with a consideration that some flow remains in the river
at any given time. This general approach to include perennial streams has been applied in this
report and no special consideration is given to intermittent streams.
Finally, the requirements for environmental flow standards can be quite variable depending on
the planned (or current) anthropogenic developments in various rivers. For example, different
environmental flow "rules" would need to apply in cases where planned water use is related to:
i) consumptive use requiring only abstraction of water; and ii) releases of water from
impoundments leading to augmentation of water. Hydropeaking (defined as an artificial rapid
and frequent alteration in water discharge) can be considered to be another special case of flow
augmentation that requires special consideration for environmental flow standards. Therefore,
the requirements and design of an environmental flow framework in the specific case with
frequent hydropeaking was considered to be beyond the scope of this report.
For a more thorough review of environmental flow literature, the reader is advised to exhaust
the following data sources;
1) The Instream Flow Council (IFC) website (www.instreamflowcouncil.org). The IFC is
an organization that represents the interests of state and provincial fish and wildlife
management agencies in the United States and Canada dedicated to improving the
effectiveness of their instream flow programs.
2) The eFlowNet website (www.eflownet.org). The eFlowNet is an independent network
whose overall goal is to integrate environmental flows into standard practices for the
management and use of river basins.
2. CONFLICTING TERMINOLOGY OF "ENVIRONMENTAL FLOWS"
The science of flow management in the context of the protection of natural ecosystems is
necessarily complex, and this is reflected in the plethora of terminology surrounding even its
most basic concepts. It could be argued that this has been a significant obstacle to the
development of an overarching framework for flow management: while practitioners struggle
with definitions of basic concepts, then progress towards unifying framework will be impeded.
The cross-cutting nature of flow management which brings together scientists from different
disciplines is one cause of the current "etymological dissonance". Here we sidestep this
problem by not reviewing all variants of terminology, but rather provide a definition and
comparison of the commonly used terms that are currently used.
Instream Flow Need / Requirement
This term is defined by the Instream Flow Council (Annear et al. 2004) as: ”The amount of water
flowing through a natural stream course that is needed to sustain, rehabilitate, or restore the
ecological functions of a stream in terms of hydrology, geomorphology, biology, water quality,
and connectivity at a particular level.". The term was popular early in the fisheries literature
(e.g. Annear et al. 2004). One key element when instream flow is invoked is the requirement to
integrate the water use needs among all stakeholders. This includes ecosystem services
provided to people living within the watershed. Some criticism of the term is related to the use of
wording "instream": it is widely recognized that the river ecosystem form and function is also
dependent on the lateral connectivity onto the floodplain, and "instream" could be interpreted as
if the floodplain is excluded from the consideration.
The term is defined by the Brisbane Declaration (2007) as: "Environmental flow describes the
quantity, quality and timing of water flows required to sustain freshwater ecosystems and the
human livelihoods and well-being that depend on these ecosystems". The term was endorsed at
International River Symposium (held in Brisbane, Australia, in 2007) by more than 750
delegates from 50 nations. It appears the term provides the most inclusive definition for the
science of flow management in the context of the protection of natural ecosystems and water
use needs among all stakeholders.
This term is used within the context of the present document.
Ecological flows are intended to indicate flows required to maintain the character of an
ecosystem. Implicit in the definition is the knowledge of what constitutes a freshwater
ecosystem in its natural state and therefore, the ecological focus of the term varies in practice.
Overall, the term is conceptually more limited than the terms above because stakeholder
requirements are excluded from this definition. The term is well defined by the New Zealand
Ministry for the Environment (2008): “The flows and water levels required in a water body to
provide for the ecological function of the flora and fauna present within that water body and its
margins”. By this definition, ecological flows are considered a component of overall
environmental flow (New Zealand Ministry for the Environment 2008).
Base flow which is often confused among the terms 'minimum flow', 'low flow’, 'low-water
discharge', 'base runoff', and 'fair-weather runoff') has been the focused target of early “instream
flow” science. The current science opinion is much more complicated in relation to hydrological
regimes as we describe herein. This term is correctly defined by the United States Geological
Survey (2005): "That part of the stream discharge that is sustained primarily from groundwater
discharge. It is not attributable to direct runoff from precipitation or melting snow."
Due to the reviewing nature of the current report, we further note that some terminology related
to environmental flows is maintained as it was originally used (i.e. in the source literature) when
describing certain assessment methods or concepts (i.e. to not confuse a method to another by
changing the terminology). It seems likely, therefore, that some historical baggage is carried
forward until the terminology becomes established as the environmental flow science matures
as a discipline.
3. REVIEW OF ENVIRONMENTAL FLOW ASSESSMENT METHODOLOGY
Flow management is an ecological imperative, and this is reflected in a vast array of
assessment methodologies internationally. Tharme (2003) described over 200 individual
environmental flow assessment methodologies and classified these techniques into four general
categories; 1) Hydrological, 2) Hydraulic rating, 3) Habitat simulation and 4) Holistic
methodologies. In order to avoid further complicating the terminology within environmental flow
assessment methodology literature (see above), Tharme's (2003) four-category classification is
used throughout this review, and the necessary use of divergent terminology within these
general categories for the purposes of review is highlighted where appropriate.
Commonly-used methods within each category are described in detail below, where they are
associated with their primary reference (Table 1). To contrast the pros and cons of each of the
four methodology categories, their specific attributes (i.e. purpose, scale, scope, duration of
assessment, relative cost and use) are listed in Table 2.
Tharme's (2003) four categories differ considerably, based on differing viewpoints regarding
how to sustain the biotic integrity of rivers. Specifically, hydrological and hydraulic rating
categories assume that a reduction in water availability will also reduce available habitat and/or
impair ecosystem function, while the habitat simulation techniques suggest that there is an
"optimum" flow where the ecosystem function is sustained (Jowett 1997). The conceptual
difference in biotic response to a flow change between these three methodological categories is
visualized in Figure 2. Holistic methods suggest that the environmental flow regimes are best
designed so that the altered flow regime follows the variability in the natural hydrograph (see
details in Section 3.4)
3.1 HYDROLOGICAL METHODS
Hydrological methods are based on analysis of historic (existing or simulated) streamflow data,
do not operate at a species-specific level, and provide an overall flow level that aims to
conserve the biotic integrity of a stream. This is based on the general assumption that more
water provides the best insurance for river biota (to a point), and sustaining some low threshold
reduces risk to the biota.
Hydrologically-based methods are still the most widely used approaches internationally
(Tharme 2003) most probably because of their ease of use and low cost, i.e., the methods use
stream flow data series, real or simulated, and don‘t require field visits. Hydrological methods
are also referred to as "historic flow" or "discharge" methods (Jowett 1997, Caissie and El-Jabi
2003), "look-up table" and "desk-top" analysis (Acreman and Dunbar 2004), "fixed-percentage"
methodologies (Tharme 2003) or "office" methods (Wesche and Rechard 1980). Quantitative
comparisons of common methods were made by Caissie and El-Jabi (1995) and Caissie et al.
(2007). In the next section, we describe some commonly used hydrological methods in more
3.1.1 The Tennant method and its derivatives
The Tennant method (also called the "Montana" method) assumes that some proportion of the
average annual flow (AAF; synonymous with mean annual flow - MAF - which is used hereafter)
is required to sustain the biological integrity of a river ecosystem. Based on original field data
collected from 11 rivers (58 cross sections, 38 different flows) in Montana, Nebraska and
Wyoming and further supplemented with additional data from hundreds of gauged flow
regimens in 21 states, Tennant (1976) recommended percentage values of MAF predicted to
sustain predefined ecosystem attributes (Table 3). Specifically, 10% of the MAF was
considered to be the lowest instantaneous flow to sustain short-term survival of aquatic life
while > 30% MAF was considered to provide flows where the biological integrity of the river
ecosystem as a whole was sustained.
Tennant (1976) recommended the proportionate flow of MAF for "low" and "high" flow periods,
or October-March and April-September, respectively, for the region where the method was
developed (i.e., North-Central USA). In other regions, temporal matching or seasonal
adjustments of low and high flow periods have been assessed, e.g., Orth and Maughan (1981)
adjusted the low flow period (10 % of the MAF) to be July-December in Oklahoma, USA.
Variations of the original Tennant thresholds are also used in other jurisdictions, e.g., 25% MAF
is regularly used as the minimum flow level required to maintain aquatic life across the Atlantic
provinces of Canada (Caissie and El-Jabi 1995).
Another modification of the Tennant method is to use a more frequent time step. Tessman
(1980) recommended a monthly time step to determine the flow thresholds, i.e., Mean Monthly
Flow (MMF). The Tessman rule recommends minimum flow guidelines as follows:
1) MMF, if MMF < 40 % MAF;
2) 40 % of MAF, if 40 % MAF < MMF < 100% MAF; and,
3) 40 % of MMF, if MMF > MAF.
The Tessman rule has been applied in Manitoba, Canada for use in perennial streams (Anon.
2007). Additional derived methods use a median monthly (or annual) discharge instead of the
MMF or MAF, e.g., the Texas Method and Lyons Method (Anon. 2005). Many other locally
used modifications have been described, and mostly to better accommodate the variations in
hydrologic regime in various geographic areas (Tharme 2003).
Mann (2006) tested the validity of the original Tennant predictions in seven western states of
the USA, including the region where the original data was collected. Mann (2006) concluded
that the Tennant's original dataset was most applicable in low gradient streams <1%, but not
representative of high gradient streams in the west (>1 %).
3.1.2 Flow Duration (Exceedence) curves, and statistical low-flow frequency methods
Flow duration curve methods define the proportion of time a certain flow threshold level is
equaled or exceeded in the particular river or region. The duration curve is calculated based on
multiple years of data, preferably using records >20 years (Caissie et al. 2007). Flow thresholds
can then be extracted from field data (filtered by expert opinion) describing the flow levels that
are required to support biotic integrity. Typically, the indices based on flow duration curves are
referred to using a Qx notation, where the subscript x indicates the exceedence percentile (or in
some cases amount of time in days, see below). For example, Q95 refers to a relatively low flow
level (i.e. flow that is exceeded 95% of the time) and Q50 to a much higher flow level (i.e. flow
exceeded 50% of the time).
The Q50 method, or median monthly flow method, was developed by the New England U.S. Fish
and Wildlife Service for catchments with good hydrological records (USFWS 1981). For smaller
ungauged catchments, an Aquatic Base Flow (ABF) was suggested and was described as the
median flow in August. The scientific support for selecting the August flow was based on the
hydrological regime in the New England region because it typically is the month with lowest flow
of the year with high water temperatures. Currently, the New England ABF method is used on a
more seasonal basis than the previous August Q50. For example in the state of Maine, the
"Seasonal ABF" is determined as the median flow for six different time periods or "seasons";
(1) Winter (January 1 to March 15): a flow equal to the February Q50
(2) Spring (March 16 to May 15): a flow equal to the April Q50
(3) Early summer (May 16 to June 30): a flow equal to the June Q50
(4) Summer (July 1 to September 15): a flow equal to the August Q50
(5) Fall (September 16 to November 15): a flow equal to the October Q50.
(6) Early winter (November 16 to December 31): a flow equal to the December Q50
It has to be noted that the above ABF flows do not describe the environmental flow that needs
to be maintained in different rivers in Maine but indicate the level of flow when no further
abstraction of water is allowed (so called "hands-off flow"; Maine Dep. 2006). The actual
environmental flow is determined using a percentage of natural flow and the allowed withdrawal
percentage varies between different stream condition classes (ranging from best condition of
grade AA to the most altered grade C class). The Q50 method has been applied also in some
Atlantic Canadian provinces, with the peculiarity that certain proportion (e.g. 70%) of the Q50
has been proposed as the minimum environmental flow (see Section 4). The scientific basis for
the use of some proportion of Q50 indices remains unproven (D. Caissie, Fisheries and Oceans
Canada, pers. comm.).
Other commonly used exceedence indices include much lower percentiles of the long-term
exceedence curve, most commonly the Q95 and Q90, which are typically used on a monthly time-
step (Caissie et al. 2007). These metrics, like the ABF, are often used to indicate a flow level
when all the abstraction should be ceased or to mark a benchmark where different level of
abstraction is allowed. In the UK, for example, the Q95 has been proposed as a threshold when
water withdrawal should be ceased or much reduced, depending on river type (Acreman and
Ferguson 2010). The Q95 and Q90 have been criticised as not providing an adequate level of
ecosystem protection for rivers (Caissie and El-Jabi 1995, Annear et al. 2004). For example,
salmonid growth rates have been shown to be much reduced at flows exceeding the Q95
(Armstrong and Nislow 2012).
It is also worth noting that in some jurisdictions, the Qx notation is applied to a flow that is
exceeded by a certain number of days of the year, for example, the Q330, Q355 and Q364 indices
used in the Czech Republic and Slovakia (see Appendix 2).
Another commonly used hydrological index that is used to determine minimum flow thresholds
is the 7Q10 (and the 7Q2 variant), based on a statistical flow-frequency analysis. The 7Q10 is
calculated as the lowest flow for seven consecutive days within a 10-year return period (and
with a 2-year return period for 7Q2; e.g. Caissie et al. 2007). Although the 7Q10 was originally
designed to protect water quality under the USA Federal Clean Water Act (Richter et al. 2011) it
is errantly used in some places to make inference on environmental flow assessment, i.e.water
quantity, (Bradford 2008). The 7Q10 is commonly used in Brazil (Tharme 2003) and at one time
was widely used in the eastern USA (Richter et al. 2011) while the 7Q2 has been mostly
applied in Quebec (Belzile et al. 1997; Caissie et al. 2007). The 7Q2 methods results in
somewhat higher flow thresholds than the 7Q10 because of the 2-year rather than 10-year
recurrence interval (Caissie et al. 2007). In Quebec's rivers, 7Q2 represents approximately 33%
of the MAF (Caissie and El-Jabi 2003).
Although frequently used, the 7Q10 and 7Q2 approaches have been strongly criticized as
lacking any scientific support for their use in setting environmental flow standards for fisheries,
and could lead to severe degradation of fishery resources (Annear et al. 2004; Caissie et al.
2007). When rigorous statistical testing was carried out comparing a variety of hydrological
methods, the 7Q10 and 7Q2 consistently produced the lowest instream flows (Caissie et al.
2007). Therefore, their use for fisheries protection is not appropriate.
3.1.3 Indicators of Hydrologic Alteration (IHA) and the Range of Variability Approach
The use of quantitative hydrological variables to support the development of ecologically-sound
environmental flow strategies is well accepted in the scientific literature (see review by Poff and
Zimmerman 2010). Developed by Richter et al. (1996), the Indicators of Hydrologic Alteration
(IHA) represent a subset of 33 ecologically-important hydrological parameters based on
variability of the annual flow regime e.g., magnitude and frequency (Olden and Poff 2003).
Calculated from daily flow data using the Nature Conservancy‘s IHA software
(http://conserveonline.org/workspaces/iha), the parameters quantify the magnitude (size),
frequency, timing, duration, and rate of change/flashiness of the annual flow regime (Table 4).
The IHA software also calculates an additional 34 parameters for five different types of
Environmental Flow Components (EFCs), namely low flows, extreme low flows, high flow
pulses, small floods, and large floods. The IHA variables can be used for long-term trend
analysis or for comparative statistical analysis to quantify change pre- versus post-activity,
e.g. dam construction or water abstraction.
The Range of Variability Approach (RVA) identifies flow targets as ranges for each of the IHA
variables (Richter et al. 1997, 1998). The RVA analysis divides each IHA variable under natural
flow (or before a change in water use) into three categories (low, middle, and high).
Boundaries can be percentile values (non-parametric approach) or the 1-2 standard deviations
from the mean (parametric approach). The Nature Conservancy recommends using the non-
parametric statistics because an equal number of pre-impact values will fall into each category
in most analyses allowing for easier understanding and interpretation. Ideally, RVA is based on
20+ years of daily hydrological data because this amount of data is required to capture the
natural variability of a system (e.g. Kennard et al. 2010). In addition to change in IHA variables,
a Hydrological Alteration (HA) factor can also be calculated for the three categories for each of
the IHA variables (Equation 1). A positive HA value indicates that the frequency of values in the
category has increased in the test period and vice versa.
Equation 1: Hydrological Alteration factor as calculated in the IHA software
Another diagnostic tool to support environmental flow assessments has been developed in
Ontario (Streamflow Analysis and Assessment Software (SAAS);
http://people.trentu.ca/rmetcalfe/SAAS.html). The distinct difference with IHA/RVA is the ability
to analyse hourly flow records, making SAAS suited for assessing large changes in flow
regimes over very short temporal scales (i.e. hourly) similar to those observed downstream of
hydropeaking facilities. (R. Metcalfe, Ontario Ministry of Natural Resources, pers. comm.).
3.1.4. Percentage of Flow (POF) methods, Sustainability Boundary Approach (SBA) and
Percentage of flow (POF) methods define environmental flows in terms of the proportion of
natural flow which can be abstracted instantaneously without compromising ecosystem
integrity. POF methods have been increasingly used to define regional environmental flow
regimes in lieu of more detailed methods and various proportions of natural flow have been
suggested depending on different river classification criteria (see section 4 for examples
internationally and in Canada). The POF approach recognizes the importance of natural intra-
and inter-annual flow variability and the concept is easy to comprehend and implement.
Environmental flow levels that are determined using % MAF methods are tied to historic flow
levels and become fixed for the incremental time period (week, month, season). Thus,
interannual variability is effectively removed. POF methods, however, use the percentage of the
flow that is currently experienced and therefore, the nuances between dry and wet years are
incorporated into the flow regime. Such capacity is important if flow regimes change (e.g. due to
the global climate change).
Building on the POF concept, Sustainability Boundary Approach (SBA) defines the extent to
which changes in the natural hydrograph can occur without impairing the flow-dependent
ecosystem benefits (Richter 2010). Thus, the SBA is a framework and not a method to
determine environmental flows; the original intent was that the boundaries within which the flow
should remain would have to be determined using other environmental flow assessment
techniques (but see below for Richter et al. 2011).
The basic approach in SBA is to use the natural hydrograph as a basis and then modify the flow
by a certain percentage of allowable augmentation or depletion of the flow (Figure 4A). The
percentage of allowable alteration can be variable throughout the year, as determined by
applying some combination of environmental flow assessment methods and further dialogue
between stakeholders and water managers. In essence, the framework does not prescribe a
certain volume of flow that is to be maintained, but an allowable deviation (expressed as a %)
from the natural condition is used instead (Richter 2010).
Some of the key benefits of this framework are (Richter 2010):
1) The natural hydrograph is maintained
2) Guidelines are set for both low and high flows
3) Scientific uncertainty is reduced as only the magnitude of flow is altered leaving
many other flow characteristics unchanged (e.g. timing, duration, frequency of
4) Easy to implement and comprehend
Soon after the SBA was published (Richter 2010), a new environmental flow framework, the
"Ecological Limits of Hydrologic Alteration" was suggested (ELOHA; Poff et al. 2010; see
description of the ELOHA framework in section 3.4). While a promising approach, some
difficulties have already emerged in application of the ELOHA in some jurisdictions, mostly
related to the implementation cost of the ELOHA framework (ranging from $100k to $2M to
develop relationships between hydrologic patterns and ecological response; Richter et al. 2011)
or time (i.e. need for an instant remedy). Thus, building upon the SBA, Richter et al. (2011)
suggested a coarse guideline to provide an interim protection of river flows, termed a
Presumptive Standard, that should be used "until ELOHA or some variation can be applied".
After a review of case studies in the USA and UK (see Richter et al. 2011), the presumptive
standard (Figure 4B) suggests that;
1) a high level of ecological protection is provided when flow alterations are within 10%
of the natural flow
2) a moderate level of protection is provided when daily flow alterations are with in 10-
3) moderate to major changes in riverine ecosystem are to be expected if alterations
are > 20% of the natural flow, with an increasing risk for alterations with a higher
deviation from the daily natural flows.
The guideline is considered to be conservative and precautionary (Richter et al. 2011).
However, authors remind that the standard may be insufficient to protect the riverine ecosystem
in hydropeaking facilities where more specific guidelines should be applied. In addition,
minimum flow levels when all water abstraction should stop may be required to the above
standards during the low-flow periods as indicated by some case studies reviewed in Richter et
As the presumptive standard is a relatively new suggestion, it is not currently known if the
standard has been adopted in some jurisdictions.
3.1.5 Strengths, weaknesses and data requirements of hydrological methods
Provided that a record of hydrological record can be obtained for a number of years, the
hydrological methods are the simplest, quickest and most inexpensive way to provide
information on threshold flow levels, but by themselves they do not produce credible flow
regimes that mimic the natural hydrograph. They may be used with other methods, however, as
part of a methodological approach to generate reasonably natural hydrographs. Hydrological
methods do not necessarily require as much fieldwork as other methods.
Despite the widespread use of various hydrological methods, quantitative studies comparing
the different hydrological methods are surprisingly few. However, Caissie and El-Jabi (1995)
and Caissie et al. (2007) provide useful comparisons of commonly used methods in a Canadian
context. Specifically, Caissie and El-Jabi (1995) compared two exceedence methods (Q50 in the
form of ABF and Q90), two Tennant derivatives (25% MAF and Tennant's "excellent" scenario;
Table 3) and 7Q10 in 70 rivers in Atlantic Canada. They concluded that Q90 and 7Q10 methods
lead to low flow predictions that could have serious adverse consequences (Caissie and El-Jabi
1995) and arrived at the same conclusion for 7Q2 in a later study (Caissie et al. 2007). The Q50
method applied at a monthly time step was recommended to be used in watersheds equipped
with a stream gauge and 25% MAF was to be used in ungauged watersheds (Caissie and El-
Jabi 1995). In a later study using a jackknife resampling technique, Caissie et al. (2007) re-
iterated some of the above conclusions and noted that of the simple hydrological methods, 25%
MAF is best suited for regional instream flow recommendations. This was stemming from the
observation that the length of the hydrological dataset had a great effect on variability of
instream flow estimates, and generally, 25% MAF showed lower variability than the other
methods even when the length of the data record was only 10-20 years (Caissie et al. 2007).
In some settings, hydrological methods have been suggested to be used at the planning level
or to set up preliminary flow targets in low risk, low controversy situations but are not
recommended for studies requiring a high level of detail (Tharme 2003; Acreman and Dunbar
2004). Lately some instream flow specialists recommend against offering preliminary flow
recommendations as they often become institutionalized by the regulators and are difficult to
recant or replace (Tom Annear, Wyoming Game and Fish Department, pers. comm.; Allan
Locke, Alberta Fish and Wildlife Division, pers. comm). Recently, hydrological methods have
been used to set the most general level of flow protection in hierarchical or tiered environmental
flow frameworks. Current scientific understanding in setting general environmental flow
standards emerges to be to use a combination of 1) percent MAF or exceedence curve
methods (or other static methods mentioned below) to determine a flow threshold level when
ALL abstraction has to stop (i.e. the so-called "hands-off flows") and 2) POF method which is
used to determine the actual percent of the current flow that can be abstracted each day (see
Section 4 for examples).
Because hydrological methods are easy to use, these methods should always be used to check
the suggested environmental flow regimes derived using other assessment (e.g. habitat
suitability) methods as an increased safety measure or a benchmark (Caissie and El-Jabi
2003). For example, if habitat suitability methods suggest very low flow levels that are, for
example, much lower than 10% MAF or Q95 for the river being considered, it would be advisable
to re-analyse the habitat simulation data in a context of overall riverine health instead of the
theoretical well-being of some selected target species. Hydrological methods may also be used
as an input within holistic framework assessments (see section 3.4).
Hydrological methods have been criticized for their lack of ecological validity and high
uncertainty with regard to hydrology-ecology relationship (Acreman and Dunbar 2004). If flow-
ecology relationships are not known for the type of river under consideration for flow
modifications, rendering a rule based on hydrological methods will be "a shot in the dark". Many
hydrological methods lead to stable (i.e. flat-lined) environmental flow regime, which is known to
lead into degradation over time (Poff et al. 1997; see also the criticism of habitat simulation
techniques in Section 3.3.1). If hydrological methods are used for flow recommendations,
appropriate validation in the target region must be carried out, and different flows should be
assigned at different times of the year (and even between years; see e.g. Alfredsen et al. 2011)
to mimic the natural hydrograph, and to better accommodate seasonal biological needs
3.2 HYDRAULIC RATING METHODS
3.2.1 General description and common methods
The hydraulic rating methods (also known as habitat retention or hydraulic geometry methods;
Tharme 2003, Moyle et al. 2011) are based on a relationship between some hydraulic measure
of a river (usually wetted perimeter or depth) and discharge (e.g. Jowett 1997). Leopold and
Maddock (1953) described simple power functions (e.g. wetted width = aQb, where Q is the
discharge, and a and b are constants) that can be used in describing changes in hydraulic
variables as a function of discharge. The constants and the exponents in these equations
should be empirically developed for each river or region, as the general form of river channels is
The methods next assume that the hydraulic measure is directly or indirectly related to habitat
quantity for a target species, almost exclusively fish (e.g. Bovee 1982, Reiser et al. 1989) or in
some instances the ecological function of the river (e.g. Gippel and Stewardson 1998). For
example, depth will determine fish presence because of body size and wetted area will affect
primary and secondary production. The most commonly used method is the ―wetted perimeter
method” that predicts wetted area of a cross-section as a function of discharge at a location
(one point) in the river (Tharme 2003). It has been frequently used in the USA (Reiser et al.
1989) and Canada (Kilgour et al. 2005).
The task, therefore, becomes to establish a relationship between the river discharge and
typically, the amount of wetted perimeter and then use this relationship to identify a "break-
point"; this is, finding a discharge below which a drastically increasing amount of river bed
becomes exposed (Figure 5). In a typical case, the response in hydraulic variable is measured
across a single or a number of "representative" cross-sections of the river channel across a
range of different discharges (or is simulated using a 1D hydrodynamic model).
One problem in defining the break point from a wetted perimeter - discharge graph is the fact
that the break-point in the curve is strongly dependent on the scale of the axis used to graph
the data (Figure 5; Gippel and Stewardson 1998; Annear and Conder 1984). For a remedy,
they showed that mathematical methods can be used to determine the critical discharge or the
breakpoint in the relationship, and suggested two methods (a Slope method and a Curvature
method) to define the break-point in the shape of the curve (Gippel and Stewardson 1998).
Shang (2008) further evaluated these two mathematical methods and concluded that the two
methods lead to inconsistent values for minimum environmental flows. The recommendation
was to not use the Curvature method but instead the Slope method with unity slope or an Ideal
Point Method was recommended (Shang 2008). Using these two methods, the minimum
environmental flow recommendations resulted in 21% of MAF in a case study river in China
(Shang 2008). In Minnesota, O'Shea (1995) found that the flow standards recommended based
on the wetted perimeter method corresponded to 39% to 122% of MAF, and in general, the
point of inflection decreased (expressed as % MAF) with increasing river size.
As hydraulic rating methods are highly dependent on the channel form, it is sometimes difficult
or impossible to find a break point that can be used to establish flow level standards (Jowett
1997). In particular, in uniform channels (consider e.g. a box shaped river bed) very small and
shallow flows can result in a very high wetted perimeter, while in reality, most of river may be
unsuitable for majority of biota (Jowett 1997) and is prone for other cascading effects like
increase in water temperature or harsh ice conditions greatly affecting the suitability of the
habitat. In rivers with triangular channel geometry, it is often impossible to find an inflection
point altogether (Jowett 1997). For example, Gippel and Stewardson (1998) failed to find an
optimum environmental flow using the wetted perimeter method in a case study in Australia.
However, they found a relationship between discharge and flowing water (i.e. in contrast to just
wetted) perimeter which they further recommend to be used instead of the wetted perimeter
method (Gippel and Stewardson 1998).
3.2.2. Strengths, weaknesses and data requirements of hydraulic methods
Hydraulic rating methods require some limited field data from the target river in order to
establish the relationship between the desired hydraulic feature (e.g. wetted perimeter) and the
discharge. The data can be either collected by multiple visits to sample the hydraulic feature
across a range of discharges or it is possible to utilize a 1-D hydraulic model (e.g. R-2 cross,
Parker et al. 2004, see also Habitat simulation methods). At a minimum, the data need to be
collected at one cross-section of the river, but is often commonly measured in a number of
transects and reaches. Because of the moderate amount of associated field work, the costs of
application of hydraulic methods to set environmental flow standards are intermediate (i.e.
higher than hydrological methods, less than habitat simulation or holistic methods).
Generally, hydraulic methods are designed to be used in rivers with well defined single
channels. Method is not well suited for braided rivers as inflections point cannot typically be
found in such channels (Jowett 1997). In channels where an inflection point can be observed,
the hydraulic rating methods result in a single flow recommendation based on the inflection
point; seasonal (or monthly/daily) or species-specific adjustments cannot be accommodated
using this methodology. On the contrary, these methods never result in a zero flow
recommendation, which is theoretically possible for habitat simulation methods. Hydraulic
methods can also be easily applied in areas where historic flow records do not exist (but see
Similarly to the hydrological methods, the hydraulic methods are recommended in situations
with insufficient information of the river systems. The premise is that these methods are coarse
and should be used with caution to set a conservative protection limit (Shang 2008). While not
implicit, it is advisable to check the environmental flow recommendations derived using
hydraulic methods against recommendations using hydrological methods to ensure the
consistency and robustness of flow standards. Otherwise, it is possible to get unsustainable
recommendations that lead to flows below historical base flows (Gippel and Stewardson 1998).
The main critique of the technique is similar to the habitat simulation methodology (see below).
The assumption that the wetted perimeter per se corresponds to biological requirements of a
species is overly simplistic and perhaps flawed. Also, these methods are estimating a proxy for
the amount of physical habitat for riverine biota, and therefore, links to population abundance
cannot necessarily be drawn (habitat quantity and population abundance are two fundamentally
separate issues, as will be discussed below; see also Conder and Annear 1987). Another
criticism is related to the selection of the transect(s) where the hydraulic variable (e.g. wetted
width) is measured. Often the selected transects are subjectively determined and there is no
guarantee that the measured transects indeed are representative for the whole river or reach
(see e.g. Fonstad and Marcus 2010).
Hydraulic rating methods are generally considered as a precursor to the more detailed habitat
simulation methods (Section 3.3), and the popularity of hydraulic methods has decreased as a
result. However, Booker and Acreman (2007) suggest that the generalized measurements of
channel form and river hydraulics are a viable trade-off for the habitat simulation methods when
a basis for further flow evaluation is needed. The information derived from hydraulic methods
can also easily be used as input tools for holistic methods (see below; Tharme 2003) as is the
case for example in Alfredsen et al. (2011) in a Norwegian Building Block Methodology case
3.3 HABITAT SIMULATION METHODS
3.3.1 Habitat simulation methods at the microhabitat scale
Different from the categories mentioned above, the habitat simulation methods aim to conserve
specific and pre-selected target species for which the habitat requirements can be reasonably
estimated in the case study area or are believed to be known from previous studies elsewhere.
As mentioned above, the theory is based on the belief that there is an underlying relationship
between the level of flow and "optimum" physical habitat conditions for the target species
(Figure 2). By using simulations of the discharge conditions, the method, in its typical and
simplest form, aims to find this optimum and set a target flow (a typical recommendation
includes a static minimum flow level) such that the amount of physical habitat for the selected
group of target species does not decline beyond a subjectively determined conservation level
(see e.g. Figure 6, but also see other approaches for determining flow thresholds based on
habitat time series below).
The habitat simulation methods have become extremely popular and even a legal requirement
in many jurisdictions in North America and globally (Tharme 2003). The popularity stems from
the establishment of the Instream Flow Incremental Methodology (IFIM) framework that was
developed for assessing the effects of flow manipulation on river habitats (Bovee 1982; Bovee
et al. 1998; Stalnaker 1995). IFIM is a holistic decision-making tool that includes, among other
steps, quantifying of the incremental differences in physical habitat that result from alternative
flow regimes (Table 5; Stalnaker 1995; Hudson et al 2003). Indeed, it is the physical habitat
tool (Physical Habitat Simulation, PHABSIM; Milhous et al. 1989) that has been most widely
used when reference to incremental methodology is made and often without the rest of the IFIM
framework tools (Hudson et al. 2003; Acreman and Dunbar 2004). This has lead to the fact that
the IFIM process is often confused with PHABSIM (Hudson et al. 2003). Technically, an IFIM
study does not need to employ the PHABSIM component if it is recognized that the habitat
conditions are not the limiting factor but, for example, the issues in a particular river are related
to water quality issues.
However, as it is commonly the PHABSIM-type component (and not full IFIM) that is meant
when habitat simulation tools are discussed, the description provided below refers only to the
physical habitat simulation component of the IFIM framework.
220.127.116.11 Generic approach
Habitat simulation methods consist of two integral parts that are linked together: 1) physical, or
hydraulic modeling providing information of changes in the physical habitat as function of
discharge and 2) modeling of the biological associations with their physical environment
(assumed to be fixed across a range of discharges).
While a suite of different nuances in habitat simulation methods can be identified (details
below), the general approach to evaluate effects of flow on habitat quantity between different
habitat methods is the same.
The habitat simulation methods have a few general assumptions (e.g. Armstrong 2010):
1. Local density of individuals (with adjustments for habitat availability) reflects local habitat
quality (i.e. "preference");
2. Preference of physical habitat attributes (for each species) is constant across
3. Individuals are free (and/or willing) to move in response to change in physical habitat
Basic steps (Figure 6) of the habitat simulation methods are described as (Hudson et al. 2003):
1. "Representative" or ―critical‖ study sites are selected
2. Hydro-geomorphology of the study site(s) are surveyed and hydraulic models (see
below) are calibrated so that changes in depth and velocity can be simulated at different
streamflows (substrate size distribution is considered to remain the same across all
3. Species-habitat association models (a.k.a. abundance-environment relationships,
habitat preference or habitat suitability index) are selected from existing literature or are
developed within the study system to represent how ‗suitability‘ of a particular stream
location for a species and/or life stage is related to physical habitat variables (most often
depth, velocity, and substrate).
4. The hydraulic model is combined with either suitability or preference curve information to
simulate how "Weighted Usable Area", WUA (an index of habitat quality-quantity), varies
5. WUA-streamflow relationships for individual species and life stages are calculated.
These relations can then be used to develop a flow recommendation.
A large number of habitat models and techniques have been described, but all operate within
the ideological template as described above, and they largely share the same assumptions,
benefits and pitfalls. While the conceptual basis of the different hydraulic-habitat models is the
same, there are differences between the habitat simulation models (both in the hydraulic and
biological models) in the detailed calculations or derivation methods of various indices used in
the process (e.g. Ahmadi-Nedushan et al. 2006; Dunbar et al. 2011). Some commonly used
hydraulic-habitat models are listed in Table 3. A recent in-depth review of habitat-simulation
methods used for setting environmental river flow is provided in Dunbar et al. (2011).
18.104.22.168 Biological component
The purpose of the biological component of habitat simulation models is to describe the
physical habitat conditions where individuals are found during different life stages and/or
seasons. The traditional habitat simulation methods assume that the spatial distribution of the
individual in the river is determined by the measured physical variables independently of other
factors (it is assumed that the "other" factors are at play at the time of collecting the biological
field data and thus are taken into account). A large variety of statistical methods has been used
to analyse the species-environment relationships and include both univariate and multivariate
functions (e.g. Ahmadi-Nedushan et al. 2006; Dunbar et al. 2011).
Traditionally, the species-environment relationships are represented in a form of a Habitat
Suitability Index (HSI; e.g. Heggenes 1990). HSI is composed as a proportionate ratio between
the habitat use of the individuals (i.e. data on where the individuals are found) for the target
species and the availability (general composition of physical variables) of the habitat within the
studied site(s) and are typically scaled from 0 (unsuitable habitat) to1 (preferred habitat). For
most applications, HSI are composed separately for velocity, depth and substrate, but many
other additional variables have also been used (Dunbar et al. 2011).
The data collection for habitat use data is often a time consuming process and the methodology
to observe the habitat use of individuals varies between species and type of habitat under
study. Common methods for fish include electrofishing (Mäki-Petäys et al. 2002), direct
observations from the river banks (e.g. Heggenes et al. 1991) or using snorkeling (Keenleyside
1962) or SCUBA diving (Linnansaari et al. 2010), and various fish telemetry methods, including
use of acoustic or radio transmitters (Scruton et al. 2002) and Passive Integrated Transponder
methods (Linnansaari et al. 2009).
Habitat availability data needed to calculate the habitat suitability indices within the studied sites
are typically collected simply using cross sectional transects, and point measurements of
velocity, depth and substrate (or some other additional variables) are taken. These point data
are sometimes used as such to calculate proportions of available habitat in the study reach
(e.g. Mäki-Petäys et al. 2002) but ArcGIS and other spatial interpolation programs (e.g. Surfer)
can be used to transfer spatially explicit point data into interpolated surface data for the study
reaches (e.g. Muotka et al. 1999; Linnansaari et al. 2010). It is inevitable that the estimation of
habitat availability in a study reach introduces an error as both using proportionate point data or
upscaling from point data to surface will be associated with some interpolation error. Sheehan
and Welsh (2009) report that by sampling of 5% of the total area of the study site, the achieved
accuracies of the availability are in the range of 57-95% depending on the interpolation method
(corresponding numbers for 2.5% sampling were 49-92% with the best results observed using
natural neighbor interpolation method for both 2.5% and 5% datasets). It is also possible to
obtain the habitat availability data using a hydraulic model (see below), but because the
availability data (for calculating a HSI) is only required for the discharge conditions when the
habitat use data is collected, the static methods described above are more commonly used.
Other methods also exist, and may include collection of habitat use data and then convene an
expert panel where professional judgment is added to develop the suitability indices, like is a
common practice in Alberta (Allan Locke, Alberta Fish and Wildlife Division, pers. comm.).
In addition to the traditional univariate HSIs, various multivariate methods have been used that
are based on different generalized linear models including logistic regression methods (Guay et
al. 2000) or generalised additive models (Ahmadi-Nedushan et al. 2006 and references
therein). Another recent development in biological modelling within habitat simulation method
family is the utilization of fuzzy logic (e.g. Ahmadi-Nedushan et al. 2008) and Demonstration
Flow Assessment techniques (Railsback and Kadvany 2008). Instead of collecting in-situ based
observational data on species habitat use, the species-environment relationships are
determined based on expert opinion (Figure 6B).
Ahmadi-Nedushan et al. (2006) provides a comparison of the advantages and disadvantages of
the different statistical methods used for estimating the species-environment relationships. In
general, the species-environment relationships have often been considered to be a weak link in
the habitat simulation methods (Moyle et al. 2011; see criticism of habitat simulation methods
22.214.171.124 Hydraulic component
The purpose of the hydraulic component in the habitat simulation methods is to provide the
information how the physical environment is changing as a function of discharge. Hydraulic
modeling can range from relatively simple solutions (1D models) to very complex multi-
dimensional models (2D and 3D solutions; Dunbar et al. 2011). Common to all models is that
the wetted perimeter of the river is divided into compartments, or cells, within which each of the
physical variables is given a simulated value for each discharge (Figure 6A). The process
involves a geo-referenced survey of the bed topography including bed roughness, detailed
measurements of water velocities and depths on cross-sectional transects (or anywhere within
the study site for 2D and 3D models) and measurement of wetted widths for at least two
discharge levels for model calibration purposes. Model validation datasets are (or should be)
also collected during the field survey (velocities, depths, wetted widths; e.g. Dunbar et al. 2011)
Traditionally, the 1D models were most commonly used due to lack of computational capacity
needed for higher dimension models, however, this is no longer a limitation. In general,
multidimensional models are used to model more complex river reaches (e.g. high-gradient or
braided river reaches) because of their denser and more dynamic cell-structure. Some 2D
models have the capacity to also consider ice covered channels (e.g. River2D; Blackburn and
Steffler 2002). However, the multidimensional models require more extensive field data
collection and thus, can be more expensive to apply (see below).
Recently, Jowett and Duncan (2011) compared the practicality and accuracy of 1D and 2D
modeling approaches in New Zealand. They concluded that the difficulty in acquiring sufficient
and accurate bed topography and the skill required in calibrating 2D models is a practical
limitation to their utility, and it cannot be assumed that they are better simply because they
require more data; the time and effort required to develop a good 2D model is not warranted in
many situations (Jowett and Duncan 2011). Furthermore, the main advantage of 2D models
over 1D models is that they should provide more accurate predictions outside the calibration
range especially at high flows in braided rivers, but improved calibration and validation
techniques are required (Jowett and Duncan, 2011).
An important step when using any hydraulic model is that after the model is calibrated to
simulate the physical conditions in a study reach, the model predictions must be validated using
a separate (independent) dataset. Such validation provides the means to understand the
accuracy and associated error of the modeling output. Surprisingly, the validation is not always
carried out even in large scale applications and furthermore, the associated error is almost
never incorporated into final habitat-discharge relationships.
126.96.36.199 Combining hydraulic and biological components
Once the hydraulic model has been calibrated and the species-environment relationships have
been established, the two separate components need to be combined into composite flow-
habitat relationship (sometimes referred to as a habitat-discharge rating curve). In a case that
univariate HSIs are used, the information regarding all the suitability values of each parameter
need to be combined into one composite number in each modeled cell of the hydraulic model
(e.g. Vadas and Orth 2001). Various techniques for establishing the composite value have been
used, and include direct multiplication, arithmetic and geometric mean and lowest SI value
(reviewed in Ahmadi-Nedushan et al. 2006). In addition, individual SI values may be weighted if
some variables are deemed more important than others (e.g. Guay et al. 2000). When each cell
is assigned a composite suitability value, a Weighted Usable Area (WUA) can be calculated by
multiplying the composite suitability value by the area of each cell of the hydraulic model and by
summing the values together for the given discharge. Alternatively, a composite suitability value
within a certain range is assigned a category, typically referred to as preferred, indifferent or
neutral, and avoided (or unsuitable) condition (e.g. Linnansaari et al. 2010) and the area of cells
belonging to each category is summed separately (e.g. Forseth et al. 2009). Finally, the
summation is carried over repeatedly across the range of discharge conditions to produce the
habitat-discharge rating curves (Figure 6).
Variability in the combined modelling output stems from both the error to accurately reflect the
species-environment relationship, and from the error associated with accurately predicting the
true value of each physical variable in each cell in the hydrological models. Typically, however,
the error is not reflected in the habitat-discharge curves and the modelling output is envisioned
The habitat-discharge rating curves can be used directly to identify environmental flow
thresholds, but they can also be used to translate a flow series of a river (water per unit time)
into a habitat time series (suitable area per unit time; Bovee et al. 1982; 1998). Habitat time
series for different flow alternatives can then be compared to natural ("no flow change") habitat
time series and therefore, the consequences of each alternative flow scenario can be analysed
as proportional habitat change compared to the natural flow. Such approach is fundamentally
different from setting target flows based on habitat-rating curves; in many aspects, the
approach is similar to hydrological methods (described in section 3.1) but units have changed
from flow to habitat. The habitat time series analysis has been used in large environmental flow
projects in Canada (e.g. in the South Saskatchewan River Basin and Athabasca River in
Alberta; Clipperton et al. 2003; Paul and Locke 2009, respectively; see also below).
188.8.131.52 A Canadian Example
A recent Canadian example of an application of habitat simulation method in a high-profile case
study is the Lower Athabasca River (LAR) Phase 2 Water Management Framework which was
developed to prescribe when, and how much, water can be withdrawn from the LAR for
cumulative oil sands mining water use (Ohlson et al. 2010). The determination of the
environmental flow regime (a combination of both fixed-flow and percent-of-flow reduction
factors and an Ecological Base Flow as referred to in the project descriptions) was based on
seven evaluation criteria (Ohlson et al. 2010), one of which was the consideration fish habitat
(Paul and Locke 2009).
The LAR case study goes to show the complexity and extensive effort that is required when
habitat simulation methods are used as a tool to provide information to be used as a part of
environmental flow recommendation. In the LAR project, habitat simulations were carried out for
six fish species (considered to be characteristic, of traditional importance or abundant in the
river), each including four life stages during two different seasons resulting in a theoretical 48
different considerations for habitat requirements, of which 12 combinations were deemed not to
be applicable (e.g. "burbot / spawning / summer" is not applicable as burbot spawn in winter).
Nevertheless, habitat information (as are required for habitat simulation) was needed for 36
considerations, but could be determined for 24 combinations (Paul and Locke 2009). A
prominent problem in the LAR project was the general lack of information regarding species-
environment relationships for the juvenile life stages during winter (Paul and Locke 2009),
which also is a period when the relatively low flow conditions naturally occur, and it is
reasonable to assume that the impacts of water withdrawal on fisheries during this period may
be the largest. Studies by the Wyoming Game and Fish Department that showed losses of
juvenile trout up to 90% during the winter in tailwater habitats of dams support this tendency
(Zafft et al. 1995).
River2D hydraulic simulations were carried out in four river reaches, and as per the preamble of
the habitat simulation approach, the information based on these reaches were considered to be
representative for the whole LAR. The WUA information was calculated for the different
species-life stage-season combinations and used to develop habitat time series for natural
flows and any proposed water management alternative. Metrics for acute and chronic habitat
loss (from natural) were calculated for each combination. Because of the large number of
outputs, the calculated habitat loss for the most flow sensitive life-stages in the matrix were
considered to represent biologically-relevant habitat changes within each river segment for
each season (Franzin 2009). These habitat-loss metrics were further used, alongside of the
other six evaluation criteria, to evaluate the effects of various water management alternatives
(Ohlson et al. 2010).
3.3.2 Strengths, weaknesses and data requirements of habitat simulation methods
In many jurisdictions, habitat simulations are considered more accurate than hydrological and
hydraulic methods to determine flow thresholds levels, and habitat simulation is recommended
in high-risk projects (Hatfield et al. 2003). Habitat simulation methods and today‘s highly
developed computer platforms easily produce WUA-discharge curves and make the
interpretation of the results straight-forward. It is also possible, at least in theory, that habitat
simulations can be used to design modified flows with the goal to improve habitat conditions for
a target species because the flows are "tailored" to accommodate a particular objective (e.g.,
Hvidsten 1993; Jowett et al. 1995).
Generally, traditional habitat simulation methods require a considerable amount of field work
and expertise to collect the hydraulic and biological data. They can be time consuming and
expensive projects. There is no shortage of debate surrounding criticisms of the method, e.g.,
Lancaster and Downes (2010a, 2010b), Lamouroux et al. (2010), Hudson et al. (2003), Hatfield
et al. (2003), Moyle et al. (2011), Dunbar et al. (2011) and Armstrong (2010). The criticisms
were synthesized by Moyle et al (2011):
"Habitat association models such as PHABSIM that infer habitat quality from AERs
[Abundance-Environment Relationships] are based on outdated concepts and unsupported
assumptions, do not deal with the processes that actually control populations, and have
dubious utility for estimating the future abundance of biomass of the target organisms
(Anderson et al. 2006, Armstrong 2010, Lancaster and Downes 2010a, 2010b), especially in
response to changes in flow. Published tests claiming to show strong relationships between
populations or biomass and the habitat index estimated by PHABSIM are mostly flawed, and
better tests show weak or no relationships" (Moyle et al. 2011, p. 79).
It is very important to recognize that the habitat simulation methods estimate only the amount of
physical habitat as a function of discharge. This "optimal" or "suitable" physical habitat is
assumed to be linked to relative abundance or biomass of a species, but such an assumption is
unproven (e.g. Bradford et al. 2011). Regardless, the knowing of the physical habitat still is a
necessary step in a process to understand the potential for perpetuation of a population of
aquatic animals (Milhous 1999).
There is considerable risk associated in implementing decisions based solely or primarily on
studies of microhabitat (Hatfield et al. 2003) and thus, habitat simulations can‘t be used alone to
recommend environmental flow standards in rivers (e.g. Ohlson et al. 2010).
Finally, for habitat simulation method to be useful, it is important that confidence intervals are
reported in the flow-habitat rating curves to address the inherent uncertainty that accumulates
from the biological and the hydraulic models (Castleberry et al. 1996).
3.3.3 Generalized (statistical) habitat models
Because applications of conventional hydraulic-habitat simulation models require considerable
field effort and experience, generalized (a.k.a. statistical) hydraulic-habitat models have been
proposed as an alternative (Lamouroux and Jowett 2005). The notable difference between the
traditional habitat models and generalized habitat models is that the latter are not based on
hydraulic model, but the change in hydraulic variables with varying discharge are computed
statistically and as an average distribution at a study reach scale (Lamouroux et al. 1995,
Lamouroux 1998). The biological data input is similar to traditional habitat models (i.e. in the
form of a Habitat Suitability Index) and the data can be either collected within the studied river
or can be obtained from literature. The use of statistical habitat models require only little
experience and field effort (some measured points of velocity, depth and substrate and wetted
width are required at minimum on two different flows to establish the statistical relationships and
a "mixing parameter"; Lamouroux et al. 1995) and the model output is in a format of WUA
curves that can be interpreted similarly as the WUA curves derived from traditional habitat
simulations. However, as the distributions of habitat variables are calculated as average reach
conditions, the models are spatially non-explicit. The generalized models have been shown to
perform well outside their calibration range (Lamouroux and Jowett 2005).
Statistical habitat models (e.g. STATHAB; http://www.irstea.fr/stathab; ESTIMHAB
http://www.irstea.fr/estimhab) have been widely used in France where they were developed
(e.g. Lamouroux and Capra 2002), but also in New Zealand (Lamouroux and Jowett 2005),
Ecuador (Girard 2009), Norway (Forseth et al. 2009) and Sweden (Harby and Sundt 2007).
Recently, generalized habitat models have been used in New Zealand to develop regionalized
environmental flow regimes (Snelder et al. 2011). The method links regionalized flow duration
curves, at-station hydraulic geometry, and generalized habitat models to make assessments at
regional scale in New Zealand. The method was also used to assess the hydrological methods
that are used in New Zealand to determine minimum environmental flows and allocation limits.
Snelder et al. (2011) concluded, not surprisingly, that in a spatially variable environment,
uniform application of a hydrological rule to define minimum flows will have spatially varying
consequences for environmental protection and reliability of supply for abstractors and
recommend the use of the method based on generalized habitat models to define regionally
varying rules for minimum flows.
Generalized habitat models share essentially all the fundamental and theoretical uncertainties
and deficiencies for which other habitat modeling approaches have been criticized (except, of
course, the uncertainty related to hydraulic modeling which is replaced by uncertainty in
statistical procedures to estimate the same variables but at the reach scale). Additional
questions rise from the non-spatially explicit nature of these models. As the statistical habitat
models are very easy to use, it is relatively easy to extrapolate the calculations into range where
no field data were collected for either biological preferences or physical variables. Naturally, the
quality of such calculations needs to be validated.
It is possible that statistical habitat models yield the same level of confidence in estimating the
amount of physical habitat as a function of changing discharge for a target species than the
more traditional habitat simulation methods, but with much less effort. However, more rigorous
testing in wider geographic area is recommended before these methods are substituted for
more traditional spatially explicit habitat models.
3.3.4. Mesohabitat models
184.108.40.206 General description
Similarly to the generalized habitat models, mesohabitat models have been developed in the
past 15 years in response to some of the criticism towards the habitat simulation methodology
at the microhabitat level.
The basic idea is that instead of associating an individual of a species into a small microhabitat
unit wherein physical parameters are described in detail, the mesohabitat models try to reveal
patterns of species association at a larger, mesohabitat, or hydromorphologic unit level. In other
words, the models make a general case that mesohabitat types with some associated cover
structures (for example, "riffle", or "pool") are related to species presence/absence or
abundance. As the quantity and the spatial distribution of the mesohabitat classes change as a
function of discharge experienced, the changes will cascade (given a discharge change that is
permanent in comparison to the natural condition) into fluctuations in target species populations
and this information can then be used to make inference on habitat availability and further,
environmental flow standards.
A few modeling platforms at the mesohabitat scale have been described in primary literature
including MesoHABSIM (Parasiewicz 2001), MesoCASiMiR (Eisner et al. 2005), the Norwegian
mesohabitat classification approach (Borsanyi et al. 2004) and Rapid Habitat Mapping
(Maddock et al. 2001). The different methods vary in their field effort-intensity (Eisner et al.
2005). The MesoHABSIM platform seems to be the most commonly applied mesohabitat
approach, and it has been applied in a number of rivers, mostly in the New England region of
the US, and a few case studies are described in
http://www.mesohabsim.org/articles/articles.html. Conceptually, the different mesohabitat
models are similar in their basic approach, and the basic steps can be summarized:
1) Mesohabitats (quantity and spatial distribution) are mapped across a range of discharge
conditions (minimum of 3, typically 4 or 5; e.g. Parasiewicz 2001) along the length of the
stream or the area of management interest (Figure 7). Depending on the modelling
platform, some field measurements are also collected to ensure the classification of
2) Mesohabitat-level biological associations are quantified. These data are collected using
standard methods (e.g. electrofishing, snorkeling etc.) depending on the size of the river.
In some cases, existing datasets can be used. Binary models (presence-absence and
presence–abundance) are then used to distinguish between unsuitable vs suitable
mesohabitats and suitable vs optimal mesohabitats, respectively, for different aquatic
species or assemblages (e.g. Vezza et al. 2011).
3) Habitat–flow rating curve is constructed and reference minimum discharge can be
determined (Figure 8). Unlike the microhabitat models that typically are associated with a
hydrodynamic model making calculations of availability of physical habitat at any
discharge, the mesohabitat models resort to fitted curves (e.g. non-linear regression) to
extrapolate between discharges.
The schemes how to classify the different hydromorphologic units are variable (see Table 6 for
just two examples), but can easily be standardized for use in a regional assessment. Eisner et
al. (2005) showed how mesoscale habitat models can be prone to subjectivity in identifying
mesohabitats (i.e. two observers could map the same mesohabitat and produce a different
description), however, the mesohabitat classification can and should be based on established
criteria that can be easily measured if necessary during the mesohabitat assessment or survey,
to reduce any subjectivity that may be otherwise introduced when determining the habitat
composition (see e.g. Borsanyi et al. 2004; Table 6).
220.127.116.11 Comparison with habitat simulation methods at the microhabitat scale
Modelling at the mesohabitat scale has some apparent benefits in comparison to microhabitat
modeling (Parasiewicz 2001). Microhabitat models are often criticized because of the fact that
the individual associations to their physical habitat are measured at a such detailed scale that
the models become sampling time dependent and thus, prone for error and often biologically
unrealistic (e.g. " for a species x, areas with depth of 10-20 cm are preferable over areas with
depth 20-40 cm"). In mesohabitat approach, the associations are considered at a larger scale
that is more robust across discharge variations (e.g. "for species x, riffle type habitat is
preferable over pool type habitat"). Generally, it is believed that an individual would be found at
the same mesohabitat unit for example throughout the diurnal cycle, whereas the microhabitat
model could be sensitive to even such small-scale temporal sampling biases (Paraziewicz
Mesohabitat models are also much less resource-intensive in comparison to microhabitat
models, at least in terms of physical habitat measurements. No cross-sectional velocity-depth-
substrate (v-d-s) data are collected reducing the field effort significantly. Some mesohabitat
modeling platforms collect a limited amount of detailed data, often in a binary format or using
actual random measurements (e.g. MesoHABSIM; 7 - 30 measurements of v-d-s in each HMU;
scale dependent). The most time consuming aspect in mesohabitat modeling is the collection of
biological data in each of the mesohabitat types (e.g. electrofishing for fish, snorkeling for
mussels/fish; or standard benthic sampling methods for macroinvertebrates). According to
Parasiewicz (2007), sampling of 500 mesohabitat units equals to 25 days of fieldwork using a
three people crew.
Stemming from the reduction in the detail of the physical habitat data collection within a "study
site", the greatest advantage of the mesohabitat approaches is their ability to quickly collect
information about physical conditions from long river sections (Parasiewicz 2007). Because the
effort can be invested to examine the river at a larger spatial scale, typical mesohabitat
applications range up to tens of kilometers within a studied river (Dunbar et al. 2011). Even in
projects where environmental flows are planned at a regional scale, mesohabitat methods have
been used to survey 5% to 10% of the stream length in one day (Vezza et al. 2011). Sampling
at a whole river length removes the problem of selecting a "representative" study reach that is
an inherent problem in microhabitat scale. Even in cases where a section of a river is sampled,
the sampled sections may be used to represent a large proportion of each individual river (and
at any rate, larger section than if a microhabitat modeling approach was used) making it more
likely that the sampled section is representative of the non-sampled parts of the river. Thus,
upscaling of the physical habitat is better justified and more likely to address issues relevant at
Mesohabitat models have been criticized on the grounds that they do not rely on hydrological
modeling to make predictions of habitat availability at the discharges where direct
measurements are not carried out, but instead, the interpolation is accomplished by curve fitting
(Dunbar et al. 2011). While this makes the data gathering more laborious (i.e. data cannot be
simulated but it has to be directly collected), it also means that the extrapolation is only carried
out for discharges where the habitat association data is valid (Vezza et al. 2011). Furthermore,
mesohabitat models may perform better in environments where hydrological modeling is difficult
(e.g. small streams with high gradient, boulder rich reaches characterized by a high degree of
flow complexity; Vezza et al. 2011).
In general, however, mesohabitat models share the same conceptual deficiencies than
microhabitat models. Similarly to the microhabitat models, the mesohabitat models are intended
as a method to estimate the amount of physical habitat available as a function of flow.
Mesohabitat models, thus, share the same incapability to take biological interactions into
account or predict changes other than the amount of physical habitat as a function of discharge.
These models can be used, for example, to evaluate if flow thresholds exist where dramatic
changes in the amount of physical habitat for a target species would likely occur. However, this
does NOT mean that such identified thresholds could be used to establish a static flow regime
for a river and assume this will maintain existing or desired biotic integrity in the river; this is not
the case. Therefore, mesohabitat models can be used to determine threshold flows under or
above which no water abstraction or augmentation should be allowed; however, the method is
not designed to recommend environmental flow regimes to preserve biological integrity. Like
micro-habitat modelling approach, environmental flow regimes based on the meso-habitat
approach don't address the inter-annual flow variability needs.
For such purpose, some comparisons between mesohabitat methods and other methodology is
available. One case study compared the output from two microhabitat models to a mesohabitat
model (Parasiewicz and Walker 2007). While only the mesohabitat model correlated with fish
abundance, and the three modelling methods suggested different amounts of habitat available,
the break point in the habitat-discharge curve at the segment scale would have provided similar
recommendation to the management (note that this is not what Parasiewicz and Walker 2007
conclude from this dataset). Recently, the MesoHABSIM approach has been used to set
environmental flows at a regional scale in Italy (Vezza et al. 2011). When compared to Q95
hydrological method, the MesoHABSIM approach resulted in average environmental flows of
1.42 times Q95 (Vezza et al. 2011). Output from this study provides also a possibility of
subjective comparison of the mesohabitat approach and hydraulic method (wetted perimeter)
approach; while calculations are not provided, the general conclusion for minimum flows would
have been, in this sample case, somewhat similar (Figure 8). However, the author informs that
the result is largely co-incidental as in general the wetted perimeter in the high-gradient reaches
where the study was carried out (P. Vezza, Dept. of Land, Environment and Geo-engineering,
Politecnico di Torino, Italy, pers. comm.)
3.3.5 Bioenergetic models
Since many deficiencies have been identified in the use of habitat simulation methods,
alternatives to habitat-simulation models have been developed. These models have been
recently reviewed in the Canadian context by de Kerckhove et al. (2008).
Models based on bioenergetics have been used increasingly to predict spatial distribution of fish
in streams. The bioenergetic models are based on the estimation of the Net Energy Intake as
cost-benefit ratio of different stream locations (i.e focal points) and require the estimation of
gross energy intake (from nutrition) and the energy spent maintaining the foraging position (e.g.
Urabe et al. 2010). Conceptually, the models based on bioenergetics are more realistic (i.e.
take into account individual fitness) than the models based on physical habitat simulations.
Bioenergetic models have been shown to be able to predict profitable stream locations (Fausch
1984; Hughes and Dill 1990; Guensch et al. 2001; Hayes et al. 2007), and relative fish
abundance (Hayes et al. 2007; Jenkins and Keeley 2010). In comparison to habitat simulation
methods, the bioenergetic models have been shown to better predict fish abundance in streams
(Urabe et al. 2010).
However, bioenergetic models also make a number of dubious or even falsifiable assumptions,
(see Urabe et al. 2010 for a discussion). The fact that bioenergetic models require estimates of
drifting prey adds another level of complexity when comprising the model output and also
makes the models expensive and time consuming to use. In addition, bioenergetic models have
only been shown to be plausible for drift-feeding fish species. Application of bioenergetic
models to make predictions of profitable stream locations may be especially difficult in winter as
the foraging behaviour of many stream fish changes seasonally (e.g. Cunjak 1996; Annear et
al. 2002; Linnansaari 2009).
3.4 HOLISTIC METHODS AND OTHER ANALYTICAL FRAMEWORKS FOR
DEVELOPING ENVIRONMENTAL FLOW STANDARDS
Holistic environmental flow frameworks have been vigorously developed during the past two
decades (Tharme 2003). Holistic methods are a group of methods, or rather, environmental
flow frameworks which are based on the need to maintain some resemblance to the natural
hydrological regime in order to sustain healthy river and riparian ecosystems. Holistic methods
aim to merge human and ecosystem flow requirements into a seamless assessment framework
(Arthington 1998). Holistic frameworks integrate social, cultural and economic values within
ecosystem protection goals. Holistic methods are sometimes referred to as expert panel
approaches, where environmental flow standards are developed in a workshop setting where
river-specific data is considered by a multi-disciplinary team of experts (typical areas including
hydrology, geomorphology, water quality and various disciplines of ecology) and importantly,
other stakeholders as the basis for consensus recommendations (Arthington 1998).
Holistic methods can be categorized into two main approaches based on either a bottom-up or
top-down strategy to describe environmental flow regime (Tharme 2003). The bottom-up
procedures are based on the supposition that it is possible to prescribe the critical components
of flow regime that needs to remain in the river (Arthington 1998). In comparison, top-down
methods assume that the entire natural flow regime is ecologically important but some flow
components can be modified or removed without ecological risk (Arthington 1998).
Whether bottom-up or top-down, all holistic approaches share some common properties
regarding achievement, or maintenance, of ecological sustainability (Gippel 2005):
1) some components of the natural flow regime cannot be scaled down, and must be
retained in their entirety
2) other components of the natural flow regime can be scaled down
3) other components of the natural flow regime can be omitted altogether
4) the variability of the regulated flow regime should mimic that of the natural flow regime
Many holistic frameworks have been described; four commonly used or emerging frameworks
are reviewed herein in some detail. Arthington (1998) and Tharme (2003) provide thorough
reviews of various holistic methods.
3.4.1 Building Block Methodology (BBM)
BBM was developed in South-Africa in the early 1990s to produce rapid advice on the
environmental flow standards using limited amounts of data (Arthington 1998). BBM is based
on a prescriptive bottom-up approach, designed to construct a flow regime for maintaining a
river in a predetermined condition (Tharme and King 1998). To obtain the predetermined
condition, the following assumptions are made (Tharme and King 1998):
1. The river biota can cope with frequent, naturally-occurring low flow conditions, and
may be reliant on higher flow conditions that naturally occur at certain times (i.e.
2. Identification of the most important components, or "building blocks", of the natural
low flows and floods, and combining them as the modified flow regime, will facilitate
maintenance of the river‘s natural biota and processes.
3. Certain flows influence channel geomorphology more than others, and incorporating
such flows into the modified flow regime will aid maintenance of natural channel
structure, and diversity of physical biotopes.
There are three main parts to BBM that take place in a sequence, each of which are described
in great detail in a comprehensive BBM manual (Tharme and King 1998; King et al. 2008):
1) A comprehensive information gathering / preparatory phase
A structured set of activities is followed to collect and display the best available
information on the river for consideration by the workshop participants. The collected
information includes social use of riverine resources, flow regime evaluations (historic
and present), hydraulic analysis, geomorphology, water chemistry, groundwater and
biological surveys for vegetation, aquatic invertebrates and fish. The BBM manual
includes detailed instructions on how the data is collected for each criteria. The
information is collected in a "Starter Document" that is provided to the participants of
BBM workshop (see below).
2) BBM Workshop
The BBM workshop typically involves ~20 people comprising of water managers,
engineers and river scientists. The workshop consists of four main sessions and
typically takes 2 - 4 days to complete. The first session is a visit to the field sites that
are being considered followed by another session where all the gathered information is
presented. In the third session, the actual modified environmental flow regime is
designed based on monthly flows and special purpose flows and reported as %MAF
(Figure 9). Finally, further research needs are identified to address major uncertainty
and to improve the environmental flow regime. A technical report is produced after the
workshop that outlines the environmental flow regime and describes the reasoning for
the different flow components.
3) Follow-up activities linking the workshop with the engineering and planning concerns
Following the workshop, the flow regime described in the workshop is incorporated in a
hydrological yield analysis. This reveals whether or not the EFR can be met without
conflict with potential consumptive users. If conflicts are identified, adjustments are
made until a compromise is achieved.
The BBM method does not examine alternative flow scenarios as it is designed to build one
consensus-based flow regime that supposedly results in a predefined river condition based on
best available scientific data (Tharme 2003). Of course, the weakness of this approach is the
assumption that the experts have a comprehensive knowledge of what constitutes a critical flow
event within the river in question.
While the applications of comprehensive BBM framework can be resource intensive and time
consuming (1-2 years; Tharme 2003), the conceptual BBM model can be used in a simplified
setting in situations where considerable data on the river system already exists. For example,
Alfredsen et al. (2011) used a simplified BBM approach to identify important flow components
needed for establishing environmental flow suggestion in a Norwegian salmon river in a
situation where limited resources were available. While not self-identifying as a BBM
application, many other projects have recognized the need to identify important "building
blocks" that are needed in order to establish a functioning environmental flow regime in
modified rivers. For example, Enders et al. (2009) used a BBM-type approach to establish
guidelines for flow regulation in an eastern Canadian Atlantic salmon river. Also, some
similarities to BBM methodology can be identified in a project in South Saskatchewan River,
Alberta, where the flow regime was compiled by a technical team by considering four
ecosystem components (water quality, fish habitat, riparian vegetation and channel
maintenance). A flow regime was determined as a consensus of the technical team and the
final recommendation included adaptive management to validate the predictions of the models
similarly to the BBM approach (Clipperton et al. 2003).
3.4.2 Downstream Response to Imposed Flow Transformation (DRIFT)
The DRIFT methodology was developed from the foundations of the BBM method in South
Africa (King et al. 2003). Unlike BBM, the DRIFT methodology is a top-down, interactive,
scenario-based approach, designed for use in environmental flow negotiations (Tharme and
King 1998). The DRIFT framework is comprehensive and includes all major abiotic and biotic
components that constitute the ecosystem to be managed (King et al. 2003). The methodology
employs experienced scientists from different biophysical disciplines which include hydrology,
hydraulics, fluvial geomorphology, sedimentology, chemistry, botany and zoology (King et al.
2003). While DRIFT methodology makes extensive use of expert knowledge, the guidelines for
selecting scientific panel members for DRIFT projects are based upon the well-established
protocols of the BBM (King et al. 2008). Moreover, the roles, responsibilities and interactions of
DRIFT panel members are governed by the step-by-step procedures built into the DRIFT
methodology and the possibility for any one member to dominate the workshops or bias the
outcomes of the scenario evaluations are removed (King et al. 2003; Arthington et al. 2003).
The DRIFT framework consists of four modules (King et al. 2003; Figure 10):
1) Biophysical module:
The component is used to describe the present ecosystem condition in the river and to
collect data on all aspects of the biota so that predictions can be made how it would
change with flow changes. Data collection is multidisciplinary and includes similar
components as described for the BBM, including informed study site selection.
Analysis includes the use of 10 hydrological statistics that are used to summarize daily
flow records (sensu 32 IHA statistics), and hydraulic modelling (either 1D or 2D) that
translates flow changes into variables that are needed to evaluate the flow-related
impacts on biota (wetted width, velocity, depth etc).
2) Sociological module:
The component identifies the groups of people directly affected by flow alteration (i.e.
"Population At Risk―, defined as those people who live along the river and use its
resources for subsistence) and describes the potential social impacts.
3) Scenario development:
The different environmental flow regime scenarios are drafted (typically less than five).
Each discipline represented in the biophysical module is then assessed using direction
and severity ratings that are decided in each project and the effects of each scenario
on subsistence users are also described. The flow regime is then negotiated using the
different severity rating trade-offs in an expert workshop environment.
The component is used to calculate the costs of mitigation and compensation for
people who are directly depended on the riverine ecosystem to be affected by the
Modules 2 and 4 are omitted if the case study does not involve subsistence users (King et al.
2003) but in addition to the basic modules DRIFT framework should be run in parallel with two
other exercises which are external to it: 1) an economic assessment of the wider regional
implications of each scenario, and 2) a Public Participation Process whereby people other than
subsistence users can indicate the level of acceptability of each scenario (King et al. 2003).
Arthington et al. (2003) provide a clear description of how specific biophysical disciplines are
considered within the Biophysical module; the example is given by using a fish component for
DRIFT framework. To illustrate the extent each discipline is (or should be) represented within
the Biophysical module, the basic steps of the DRIFT-fish, as reported in Arthington et al.
(2003), are shown:
"The basic steps in the fish component of DRIFT are the following:
Step 1. Review of literature to produce a compilation of published flow-related information on
each fish species in the study rivers.
Step 2. Selection of study sites to characterize river reaches likely to be affected by existing and
future water resource developments.
Step 3. Seasonal field surveys at each site to determine fish species composition, abundance
and habitat use in relation to flow conditions.
Step 4. Analysis of field data to generate habitat preference curves for each fish species.
Step 5. Tabulation of field data and information from literature review to produce summary of
flow-related data on each fish species.
Step 6. Development of scenarios of flow regime change for evaluation using DRIFT.
Step 7. Development of protocols to document the consequences of flow regime change for
each fish species at each study site.
Step 8. Prediction of the ecological and social consequences of flow regime change for each
fish species at each study site.
Step 9. Preparation of a monitoring strategy to assess the outcomes of environmental flow
Step 10. Implementation of monitoring programme, evaluation of ecological outcomes of any
environmental flow provisions, and adjustment of those provisions in the light of new
knowledge generated by monitoring (and research)." Arthington et al. (2003), p. 643-
Considerable uncertainty in the decision-making is inevitable when the ecological
consequences are predicted for the different species within each biotic component (e.g. step 8
for fish above). The DRIFT framework accounts for this uncertainty by using "severity ratings"
and the predicted direction in change while the confidence level in all these decisions is also
reported. When the different flow scenarios are contrasted, patterns in the direction of change
and severity emerge, and can be used for making a decision between the scenarios despite the
uncertainty. Like any other holistic framework, DRIFT must also be followed up with adaptive
management that is based on quality monitoring data (Arthington et al. 2003).
The implementation costs of the DRIFT framework can be significant depending on the scope
of the research carried out in the biophysical module (Acreman and Dunbar 2004). However,
DRIFT is more suitable for trade-off negotiations than the BBM method as implications of not
meeting the environmental flow targets can be assessed (Tharme 2003).
3.4.3 Benchmarking and the derived frameworks
The idea of "benchmarking" was originally developed in Fitzroy Basin, Australia (Arthington
1998). The benchmarking methodology was designed to link information on alterations of
natural flow regimes to ecological consequences of flow regime change (Arthington and Pusey
The main idea is to evaluate the condition of a range of rivers (or river reaches) that have been
subjected to various degrees of flow regulation and water resource development (Arthington
and Pusey 2003). The flow regime of the study river is described using a set of key flow
statistics that are thought to have important ecological relevance (12 variables in the original
method; Arthington 1998). The next step is to describe the percentage of change in each flow
statistic from its natural (pre-regulation) value and link that to the observed ecological (or
geomorphologic) impact. These relationships (% change in flow statistic versus ecological
impact) can be used for making probability statements about the ecological implications of
altering a river‘s flow regime by specified amounts compared to the natural regime (Arthington
and Pusey 2003). This can be further used to establish limits, or "benchmarks", for the
maximum "allowable" change in each flow statistic in comparison to the natural condition.
The concept of benchmarking was further adopted within a new framework that was designed
to fight a " growing temptation to ignore natural system complexity in favor of simplistic, static,
environmental flow „„rules‟‟ to resolve pressing river management issues" (Arthington et al.
2006, p. 1311). They suggested a new approach that incorporates essential aspects of natural
flow variability common across particular classes of rivers that can be validated with empirical
biological data and other information in a calibration process (Arthington et al. 2006). Therefore,
the framework was designed to provide sustainable flow management criteria for larger groups
of "similar" rivers.
In this framework, four steps were suggested (Figure 11; Arthington et al. 2006):
Step 1: Develop classification for reference streams;
Groups of similar streams are defined using ecologically-relevant flow statistics that are
identified in an analysis of the respective natural hydrographs (Figure 11a)
Step 2: Develop frequency distributions selected flow variable in each class;
Rivers within a similar reference group have natural variability with respect to the
selected ecologically-relevant flow statistics. By combining the information from multiple
rivers within the same reference class, a composite hydrographs is developed and can
be used as a "norm" within each class of rivers (Figure 11b)
Step 3: Frequency distributions are compared between flow modified and natural streams within
the same class;
A measure of deviation from the norm is obtained for each ecologically-relevant flow
statistic (Figure 11c)
Step 4: Develop flow-response relationships using selected ecological health indicators from
reference and flow modified steams for each flow variable.
Threshold conditions for the selected flow variables are identified based on a number of
river health indicators (e.g. aquatic macroinvertebrate diversity, density of a select fish
species etc.). These thresholds become the "benchmarks" that are used to design an
environmental flow regime within each river class.
Arthington et al. (2006) is a clear precursor for the ELOHA framework (see below).
3.4.4 The Ecological Limits of Hydrologic Alteration (ELOHA) Approach
The ELOHA framework represents a recent consensus view from a group of internationally
recognized environmental flow scientists (Poff et al. 2010), building on a previous framework
described in Arthington et al. (2006). ELOHA can be used to determine ecological limits of flow
alteration at a regional scale and, therefore, simultaneously for a large number of "similar"
rivers (Poff et al. 2010). The ELOHA framework does not reveal any new environmental flow
assessment techniques per se but rather provides a consistent approach for analysis and
synthesis of available information (i.e. using existing hydrologic techniques and environmental
flow methods) to achieve environmental flows (Poff et al. 2010).
The goal of the process is to develop regional environmental flow standards that are rooted in a
science-based, quantified evaluation of the effects between different categories of flow
alteration and the consequent effects on riverine biota.
The ELOHA framework consists of five basic steps divided into scientific (steps 1-4) and social
processes (step 5) (Figure 12):
Step 1: Hydrologic modeling for baseline and status quo hydrographs in the region;
Step 2: Classification of rivers (or river segments) based on flow regime and geomorphic
Step 3: Determination of the extent of alteration;
Step 4: Development of flow-ecology relationships for the different river types; and
Step 5: Establishing environmental flow standards with subsequent monitoring and
Because the ELOHA framework provides a possibly promising approach to be used in the
Canadian context, each of the steps are described in more detail with reference to their current
status in Canada.
Step 1: Hydrologic modelling
The first step of the framework relies on the use of a hydrologic database which is explored to
extract daily flow hydrographs for simulated baseline and developed conditions. Baseline
conditions consist of rivers that have been minimally altered (i.e. "reference sites") whereas the
developed conditions refer to rivers where the flow regime has been altered. The classification
of the hydrographs into two categories should be carried out using a single time period (to be
able to separate human and climatic influences) of at least 10-20 years with flow record (and
the process may have to be iteratively renewed depending on the potential effects of climate
change; see below). Statistical techniques and hydrological rainfall-runoff models can be used
to generate data for ungauged locations (Poff et al. 2010, and references therein).
In Canada, the Water Survey of Canada (WSC) maintains a national database (HYDAT) for
gauged daily hydrological data (rivers, lakes and reservoirs) collected by provincial and federal
agencies. Currently, WSC releases an updated Access database (free to download)
approximately every three months containing the most recently available hydrometric data in
addition to station information. An examination of the available daily hydrometric data, the
stations and the length of available data records emphasises the variability in both spatial and
temporal coverage across Canada. Figure 13 shows that the majority of sites have 20 years or
fewer of available data.
Examination of Figure 14 demonstrates the spatial disparity of data collection with a strong
southern bias to station location. This is not unsurprising because the majority of the population
lives in southern Canada and it is easier to maintain these gauges. However, recently additional
data is being collected in northern Canada (Figure 14a).
Established by the WSC (Environment Canada), a subset of 255 hydrometric gauging stations
has been allocated to the Reference Hydrometric Basin Network (RHBN) (Figure 14b). The
RHBN stations provide a network for monitoring longer-term changes and impacts of climate.
Brimley et al. (1999) and Harvey et al. (1999) defined the original station selection criteria: (i)
sites should represent near-natural conditions with less than 10% modification from natural flow
conditions; (ii) absence of significant regulations or diversions upstream of the gauging station
(less than 5% of the area regulated); (iii) minimum of 20 years of hydrological data; (iv)
longevity of the station in its current pristine or stable state in the future; and (v) highly accurate
data records. The final criterion refers to the breadth of coverage of the different types of
available hydrometric stations (seasonal, continuous, streamflow, and lake level). Of the 255
RHBN stations, 223 record river discharge.
Coupled reference and developed conditions hydrologic time series should be developed for all
locations in the region where environmental flow protection is anticipated (Poff et al. 2010). The
hydrological data for both reference and developed are also needed to establish the flow
alteration-ecological response relationships (i.e. step 4 of the ELOHA framework). The existing
HYDAT database with the prescribed reference stations provides a good starting point for
implementing the step 1 of the ELOHA framework in Canada. However, it must also be stated
that unresolved issues remain concerning the minimum hydrometric series length required to
capture climate variability in temperature and precipitation, which drive runoff processes. This
issue is further exacerbated by the influence of anthropogenic climate change as a further
Step 2: Classification of hydrologic regimes
The rivers that have been identified to represent the reference conditions (step 1) are classified
into similar groups or "river types" (similar to the Arthington et al. (2006) framework). The region
within which the classification is spatially and temporally variable from areas defined by
jurisdictional boundaries (e.g. provinces) to more natural biophysical domains within a larger
geographic area (see below). Classifying is based on ecologically-relevant characteristics of the
flow regime and geomorphology. Identifying river types is based on flow-geomorphology
relationships and corresponds to different ecological characteristics. Therefore, the form and
the direction of an ecological response to flow alteration is hypothesised to be similar within, but
vary between, river types (Poff et al. 2010). This allows the generalization of environmental flow
thresholds (step 5) for all the rivers within the same type (Arthington et al. 2006). It is notable
that the river types not necessarily are geographically contiguous, meaning that different
segments of a river (known as "analysis nodes" in ELOHA) may be classified as different river
types; headwater tributaries and a mainstem river in the same basin would likely be classified
as different river types (Kendy 2009). This will also depend on the geomorphology component
because for example, a given flow level may have different ecological consequences in a
bedrock vs an alluvial bed river. Various methods and tools exist for both hydrological (reviewed
in Olden et al. 2011) and geomorphological classifications (e.g. Elliot and Jacobson 2006;
Thompson et al. 2001).
In the Canadian context, Monk et al. (2011) provide an initial hydrological regime classification
for rivers across the country. Using the RHBN sites with daily data available for hydrological
years 1970-2005, the authors quantified the timing (shape) of the annual hydrological regime,
which allows spatial (between-station) patterns to be examined. Based on standardised weekly
runoff averages, an agglomerative hierarchical classification method (using Ward‘s approach)
modified from Hannah et al. (2000) and Harris et al. (2000) was applied to identify homogenous
hydrological regions with similar timing of the annual flow regime, regardless of absolute
magnitude. Six hydrological regime shape classes were identified reflecting known
geographical and climatic variability across Canada (Figure 15). Following part of the
standardised approach to classification proposed by Olden et al. (2011), the classification
approach adopted by Monk et al. (2011) provides a starting point for the development of an
ELOHA framework in Canada for future environmental flows assessment. The method could be
expanded to include non-RHBN stations in addition to developing a method to identify the river
type for hydrologically-impacted hydrometric stations (to be used in step 3) or non-gauged
Step 3: Determination of the extent of alteration
In the third step of the ELOHA framework the extent of flow alteration is determined as the
deviation of the flows between the historic (baseline) and the current day condition in each
individual river (or analysis node) for each hydrological metric and is expressed as % of
deviation from the baseline (Poff et al. 2010). These data are used as an input when developing
flow alteration-ecological response relationships (step 4). A variety of tools exists for
calculating the extent of flow alteration; IHA and SAAS are described above (Section 3.1.3) in
addition to another commonly used software the Hydrologic Alteration Tool (HAT) of the U.S.
Geological Survey‘s Hydroecological Integrity Process (HIP) package is described in Henriksen
et al. (2006).
The key part of the ELOHA framework is carried out in step four where biological and
hydrological data are combined into flow alteration-ecological response relationships (flow-
ecology relationship for short), similarly to the framework described in Arthington et al. (2006,
Figure 12). The process of developing flow-ecology relationships begins by formulating testable
hypotheses how flow is expected to alter the ecological response variables that are chosen for
evaluation. The hypothesis formulation is carried out as a collaborative process by scientists
familiar with the ecology and hydrology of the region that is being considered (e.g. Haney et al.
Step 4: Development of flow-ecology relationships
The next stage is to collate existing ecological data to quantify the relationships. The data
needs to be collected within each river type identified at step 2 of the framework because
globally transferable flow-ecology relationships have not yet been developed (Poff and
Zimmerman 2010). The ecological response variables ("health indicators" in Arthington et al.
2006) that are used to establish flow-ecology relationships should be: 1) sensitive to existing or
proposed flow alterations; 2) can be validated with monitoring data; and 3) are valued by society
(Poff et al. 2010). The response variables can be simple metrics (e.g. presence-absence of a
species, changes in relative abundance of a species) or based on composite indices. A number
of different indices have been developed for fish (e.g. Fausch et al. 1984; Pirhalla 2004; Bain
and Meixler 2008; examples from the USA) and lotic macroinvertebrates (e.g. Extence et al.
1999; Monk et al. 2007; Armanini et al. 2011a; examples from the UK and Canada).
While no comparable fish metric for the development of flow-ecology relationships currently
exists in Canada, a composite index for macroinvertebrates has been developed. The Canadian
Ecological Flow Index (CEFI) was developed using paired benthic macroinvertebrate
community samples and flow velocity data extracted from the Canadian Aquatic Biomonitoring
Network (CABIN) database (Armanini et al. 2011a). The index approach summarises flow
velocity preferences for common benthic macroinvertebrate taxa within a sample. Therefore,
CEFI offers a quantitative method for linking hydrological regime variability (both natural and
anthropogenic) to compositional variation in the ecological community. To test CEFI and its
response to hydrological variability, Armanini et al. (2011b) demonstrated how runoff regime
type (using the classification of Monk et al. (2011) influenced CEFI community response. In
addition, Peters et al. (2012) highlighted a practical approach to utilise CEFI in an
environmental flows assessment. Using a reference condition approach, observed CEFI values
for potentially impacted sites can be compared with an expected value for regional
hydrologically-similar rivers. The index values have not yet been collected across a gradient of
flow-altered sites in order to establish a flow-ecology relationship. Nevertheless, CEFI seems as
a good candidate for initiating such undertaking in a nationally standardized context.
In ideal situation, the flow-ecology relationships are quantified as % change in health indicator
vs % change in flow metric (Arthington et al. 2006) but may have to be expressed in a more
simple format such as categorical or binomial relationships (Poff et al. 2010). In some cases,
data are abundant to develop flow-ecology relationships across a gradient of flow alterations
(Apse et al. 2008). More often, this may not be the case. In cases of limited data, different
approaches have been used to advance in the ELOHA framework. Different studies have
resorted to expert opinion (Haney et al. 2008) and statistical analysis (Konrad et al. 2008; Webb
et al. 2010) to establish the flow-ecology relationships. Compilation of data will at any case
identify conditions where ecological data are missing; such information can be used to direct
future field studies into strategic locations (Arthington et al. 2006). Finally, whether data are
abundant or scarce, scientists must account for confounding factors such as changes in water
quality and temperature, ice regime, physical habitat degradation, and invasive species, which
may cause substantial impact even with minimal flow alteration (Kendy et al. 2009); however, it
is possible to limit the influence of confounding variables using appropriate statistical
approaches (e.g. Armanini et al. 2011a).
Step 5: Establishing environmental flow standards
Once flow-ecology relationships have been established for key hydrological variables,
environmental flow standards can be established through a stakeholder-driven process. The
benchmarking approach (Arthington et al. 2006) has been adopted as a tool to help establish
ecologically and societally acceptable flow thresholds (Poff et al. 2010). Societally acceptable
flow thresholds may be related to public concern around the "value" of each individual river (or
segment) within the river type and/or analysis node, and the framework could benefit if analysis
nodes were assigned to different "management classes" (e.g. pristine, modified, highly modified
etc.), as has been done in other approaches (see Section 4). The ecological goals and
therefore, e.g. maximum limits for water abstraction, could then be differently assigned within
the management classes (intuitively this would be stringent "rules" in pristine waters, and more
relaxed thresholds in modified catchments). The ELOHA framework does not pre-determine the
societal process by which environmental flow standards are finally derived, but case studies
show that this can take place through stakeholder/expert consultation and committee meetings
(Poff et al. 2010). Thus the ELOHA framework, like other holistic frameworks, is highly suitable
to establish an initial set for regional environmental flow standards that can be refined in an
iterative loop of adaptive management based on monitoring data (Poff et al. 2010).
Examples of ELOHA projects
A number of projects using at least some parts of the ELOHA template have recently been
initiated (described in the ELOHA website: http://conserveonline.org/workspaces/eloha).
Several projects within the United States are currently applying elements of ELOHA to
accelerate the integration of environmental flows into regional water resource planning and
management (Kendy et al. 2009). In the US, ELOHA has been applied in Pennsylvania,
Tennessee, Michigan, Arizona, Colorado and Washington at least in some parts of the state
(Apse et al. 2008; Kendy et al. 2009; Sanderson et al. 2011). ELOHA has also been used in
Australia (Arthington 2009). Many projects are at a stage where only some steps of the ELOHA
framework have been accomplished or adopted. For example, a project in Huai River basin in
China used ELOHA steps 1 to 3 to provide a foundation for development of hydro-ecological
relationships in the region (Zhang et al. 2011). In Canada, Peters et al. (2012) recommended a
framework to be used to establish environmental flows in agricultural regions of Canada with
many similarities to ELOHA.
While implementing full ELOHA framework across Canada would probably take substantial
amount of time to complete, it provides a number of benefits over more traditional methods
used for establishing environmental flow regimes. The main benefits are that environmental
flow standards are developed for large, "similar" regions simultaneously, and the decisions are
based on scientifically driven, testable, and quantified relationships that link hydrological
change to meaningful ecological variables.
3.4.5 Strengths, weaknesses and data requirements of holistic methods
Holistic methods have several advantages in comparison to other methodologies. These
methods consider all aspects of the flow regime, retain the natural-like hydrological regime and
address all relevant components of the river ecosystem and take the associated societal needs
into account (Tharme 2003). The methods, in general, operate under the premise that
ecosystem must be assessed and satisfied before humans can take water (Arthington 1998).
Holistic methods rely considerably on professional judgment and expert opinion. It can be
debated who qualifies as "an expert" and there is always a risk that expert panels are dictated
by a few dominant personalities or that decision process becomes flawed due to interpersonal
dynamics (Gippel 2005). Depending on the depth of evaluation, data collection, and the extent
of expert consultation, applications of holistic framework can be time consuming and very
expensive. Also, if single flow regime is prescribed by an expert panel, the natural inter-annual
variation in flow regime will be lost; however, recent holistic applications have recognized the
importance of inter-annual variability and describe different flow regime depending on the
"water availability" in each year (i.e. dry, normal, wet year; Alfredsen et al. 2011).
Data requirements for holistic methodologies are not easily specified. Many holistic methods
rely on expert opinion and therefore, well informed specialists within each ecosystem
component to be addressed are needed (King et al. 1999). The holistic methods can utilize data
collected using any other environmental flow assessment methodology as an input and
generally, the expert panel benefits from receiving as much supporting data as possible,
including the data need concerning recreational use and subsistence requirements of the local
people (King et al. 1999).
Data requirements for ELOHA-type framework are considered above. The intention in the
ELOHA framework is to initially invest in simple tools and rely on existing data and continue to
more complex and expensive approaches at later stages (Poff et al. 2010). This will include
additional data collection depending on where the largest data caps are identified after the initial
models have been formulated. It is possible that flow alteration - ecological relationships are
difficult to establish for some river types if the gradient in terms of flow alterations is limited (e.g.
Arctic Canada). In such cases, alternative approaches have to be considered (e.g. DRIFT
4. SUMMARY OF ENVIRONMENTAL FLOW GUIDELINES CURRENTLY USED IN
CANADA AND INTERNATIONALLY
4.1 CURRENT ENVIRONMENTAL FLOW GUIDELINES IN CANADA
The information represented herein is collected from three general sources; 1) current (i.e. Oct
2011 - Jan 2012) www-sites of the Agency/Ministry responsible for water allocation in each
province, 2) personal communications with provincial experts (Table 7) and 3) a review by
Katopodis (2009). The list of provincial contacts is intended as a source for an initial point-of-
contact in the respective provinces with regard to environmental flows and does not imply an
all-inclusive list of provincial experts. In addition, an Agency / Ministry responsible for water
allocation is reported below for each jurisdiction in order to facilitate further enquiries regarding
Some general similarities were identified between the different jurisdictions with regard to
environmental flow guidelines. The shared features can be summarized as:
All jurisdictions share a similar intent to protect water resources from anthropogenic
impacts in rivers; this is typically contained in provincial act related to water resources
("Water Rights", "Water Protection Act", "Water Act" etc.). The provincial legislation
protects water resources using relatively general description (i.e. reference to protection
of aquatic biota/ habitat), without specifying the means how to e.g. determine
environmental flow thresholds. In general, overarching rules describe that alterations
must be sustainable, should not cause any significant adverse effects to the
watercourse and that "some" amount of water is required in river to maintain a healthy
aquatic habitat. Environmental flow guidelines, if these exist, tend to be
recommendations or "best advice" to the regulator and not legally binding.
A provincial permit (e.g. "water withdrawal approval", "a watercourse and wetland
alteration permit", "permit to take water" etc.) from an agency responsible for water
resources (i.e. typically provincial government) is generally required if flow alterations
are considered. Some small amounts of water for consumptive use can be withdrawn
without a permit (typically in the range of 20 to 50 m3 / day). Many different branches of
local government are typically consulted in the permitting process. Federal agencies (i.e.
DFO) are involved in cases where a potential for "Harmful Alteration, Disruption or
Destruction of fish habitat" (i.e. HADD) needs to be considered.
In terms of HADD considerations, many provinces aim to a set of rules that can be used
to differentiate between "no HADD" and "potential HADD". This translates into a 2-tiered
structure (Figure 16) with some general guiding rule for distinguishing the "no HADD"
cases (referred to as Level 1 assessment below). In projects where the general guiding
rules cannot be met, site-specific studies are to be carried out to determine more
specific environmental rules (referred to as Level 2 assessment below)
Definitive frameworks or protocols to set environmental flow standards do not seem to
exist for cases where potential HADD is invoked (i.e. site specific regarding Level 2
assessments; but see Lewis et al. 2004 for British Columbia). The protocol is
determined by case-by-case basis, and environmental flow standards are typically
developed in a public participation decision making process.
Cumulative effects of water abstraction are very poorly managed and no central
databases seem to exist for small scale (i.e. Level 1 or "no HADD") consumptive
projects. This may possess a potential problem because a large number of small
consumptive projects may constitute a large cumulative effect.
4.1.1 British Columbia
Responsible agency: British Columbia Ministry of Environment
Level 1 assessment: The current environmental flow guideline(s) for British Columbia are
comprehensively described in Hatfield et al. (2003), and the related habitat assessment
methods in Lewis et al. (2004). Environmental flow guidelines are determined separately for
fishless and fish-bearing streams, although the maximum diversion rate is determined as the
Q80 for both stream types. For fishless streams, the environmental "cut-off" flow is determined
as the Q50 during the low flow month (the low flow month is defined as the calendar month with
the lowest median flow, based on natural mean daily flows).
The recommended environmental flow thresholds for fish-bearing streams are adjusted on a
monthly basis. The environmental flow for the lowest flow month is set as Q90, and for the
highest flow month as Q20. The environmental flow thresholds for all other months are
calculated as a percentile between Q90 and Q20 using a weighted function (Hatfield et al. 2003).
As a result more water is available for diversion during high flow months than during low flow
Some practitioners have reported that the Level 1 rules are conservative to such an extent that
they are rarely used - because many projects requiring water would be deemed not
economically viable if based on guidance of these Level 1 environmental flow rules (R. Ptolemy,
British Columbia Ministry of Environment, pers. comm.).
Level 2 assessment: Lewis et al. (2004) provides a guideline for carrying out detailed
assessments for various environmental criteria that should be used as part of the process to
determine environmental flows at a more detailed level. Habitat simulation methods are
frequently used as part of the assessment, but are not used to determine static environmental
flow levels, but the analysis is based on "building blocks" that are identified based on biological
understanding life cycle of some species that are considered to be valued for special
Responsible agency: Alberta Environment and Water
Level 1 assessment: After a comprehensive review, the province of Alberta recently established
a guideline for environmental flow needs that is to be used in the absence of having site-
specific information that could otherwise be used to establish an environmental flow (Locke and
Paul 2011). The guideline uses a combination of POF and exceedence methods. The guideline
prescribes the greater of either:
1) A 15% instantaneous reduction from natural flow or
2) The lesser of either the natural flow or the Q80 natural flow based on a weekly or
monthly (depending on the availability of hydrology data) time step.
In other words, no water abstractions are allowed for the lowest flows that occur up to 20% of
the time, and for the remaining 80% of the time, up to 15% of the natural flow can be
withdrawn. The guideline is based on the objective to fully protect the aquatic environment
relative to natural conditions. Approximately a dozen licences have been issued based on the
desktop guideline (as of Dec. 2011; A. Locke, Alberta Fish and Wildlife Division, pers. comm.).
Level 2 assessment: No guideline currently exists. Environmental flows have been recently
determined in a number of large-scale projects (e.g. Clipperton et al. 2003, Goater et al. 2007,
Ohlson et al. 2010). The approach in these projects has been holistic, using multiple
environmental criteria and a range of different methodologies for assessment. The
assessments have not, however, followed any fixed framework, but have been carried out on a
4.1.3 Saskatchewan and Manitoba
Responsible agency: Saskatchewan Watershed Authority; Manitoba Conservation and Water
Level 1 assessment: Neither province currently has public guidelines for establishing
environmental flow standards, although development for provincial guidance in underway in
both provinces. In Manitoba, the Tessman rule (see Section 3.1) has been used.
Level 2 assessment: No guideline exists, and environmental flows are determined case-by-
Responsible agency: Ministry of Natural Resources environmental flows related to hydropower
developments), Ministry of Environment (consumptive use).
Level 1 assessment: There is currently no guideline and various rules have been used.
However, both the Ministry of Natural Resources and Ministry of Environment are drafting
science advice related to environmental flows to support respective approvals processes.
Recent suggestions have considered Q70 as a general guideline in rivers where species at risk
are present, and Q75 in other cases.
Level 2 assessment: No guideline exists, and the assessment methods are chosen by the
proponents requiring flow alteration.
Responsible agency: The Le ministre du Développement durable, de l'Environnement et des
Level 1 assessment: An "ecohydrological method" (Belzile et al. 1997; Berube et al. 2002) has
been described to determine conservation flows in the rivers of southern Québec (south of 52o
N). The method has some conceptual similarities to the ELOHA framework. Different
environmental flows are set for 10 different ecohydrological regions, wherein valued target (fish)
species were evaluated. Critical life phases of the selected target species were determined and
different seasonal flow exceedence rule was applied (not based on flow-ecology relationships)
in each ecohydrological/species composition area. It is not known if the method is supported by
the responsible agency, and if any guidelines apply in northern parts of Quebec.
Level 2 assessment: There is no current guideline and the proponent of a project must justify
the chosen method that is used to establish environmental flows. Habitat simulation methods
have commonly been used to establish environmental flows on a case-by-case manner.
4.1.6 Atlantic Provinces (New Brunswick, Nova Scotia, Prince Edward Island and
Newfoundland and Labrador)
Responsible agencies: New Brunswick Department of Environment; Nova Scotia Environment
and Labour; Prince Edward Island Department of Environment, Labour and Justice;
Newfoundland and Labrador Department of Environment and Conservation
Level 1 assessment: All the Atlantic provinces have some general guideline regarding
environmental flows, although the derivation of the is not well documented. The guideline in all
Atlantic provinces consist of a rule when all water abstraction must stop. In New Brunswick and
Prince Edward Island, 70 % of the monthly Q50 is used (in PEI, another cut-off threshold is
added such that the above rule cannot lead to withdrawal below monthly Q95). In Nova Scotia,
25% MAF rule is used and in Newfoundland and Labrador, the use of "low quartile of mean
monthly lows method" has been suggested (i.e. Q25 of MMF; Rollings 2011).
Level 2 assessment: There is currently no guideline; habitat simulation methods have been
4.1.7. Eastern and Western Arctic
Responsible agencies: In Northwest Territories various land and water boards depending on the
region; Nunavut Water Board. No information was obtained for Yukon.
Level 1 assessment: In Northwest Territories, DFO drafted a guideline for winter water
withdrawal in 2005, which allowed a 5% instantaneous reduction from natural flow (Cott et al.
2005). The guideline was revised, however, in 2010 and no fixed allowable reduction was
described (DFO 2010). The current assessment is carried out on a case-by-case basis and if
withdrawal is allowed, the recommendation is typically 5 - 10 % of the instantaneous flow by the
time of withdrawal. No specific guideline exists in Nunavut.
Level 2 assessment: N/A.
4.2 ENVIRONMENTAL FLOW GUIDELINES IN OTHER SELECT COUNTRIES
Similarly to Canada, there is no nationwide framework for establishing environmental flows in
the USA and the different states describe limits to flow alteration independently. Traditionally,
habitat simulation methods have been extensively used to determine suitable environmental
flows, targeting some valued species (Tharme 2003), and this still is the preferred method in
many states. However, an increasing number of states have adopted various ways to classify
rivers based on their ecological or societal values, and establish environmental flow standards
based on some combination of hydrological methods, within the river classes or types. In many
states, these steps have some resemblance to the ELOHA framework, and a few states are
endorsing it fully (e.g. Minnesota, Pennsylvania).
Details for current environmental flow assessment methodologies for altogether 18 different US
states can be obtained by reviewing Locke and Paul (2011, p. 59-72) and the case studies in
ELOHA toolbox website [http://conserveonline.org/workspaces/eloha/documents/template-kyle].
4.2.2 European Union
The European Union member states are mandated by the EU Water Framework Directive
(WFD) to achieve good ecological status in all waterbodies by 2015 (e.g. Acreman and
Ferguson 2010). While the WFD does not implicitly make reference to environmental flows, it
is generally accepted that ecologically appropriate hydrological regimes are necessary to meet
the WFD requirements (Acreman and Ferguson 2010). The various methods to describe
environmental flows in different EU member states were reviewed prior to the 2nd Workshop on
Water Management, Water Framework Directive and Hydropower, held in Bryssels in
September 2011; Kampa et al. 2011). The review showed a very wide range of guidelines
concerning environmental flows in the EU (Appendix 2). Most countries have a recommendation
for setting environmental flow regime but some countries determine environmental flow
guidelines a case by case basis. To capture the range of different level of environmental flow
protection in the EU, three cases are highlighted.
In Norway, no nationwide environmental flow guideline has been established, and the
evaluation of flow standards is carried out on a case-by-case basis. As a general rule, the Q95 is
used and is calculated separately for summer and winter seasons. More water is required for
rivers that have a special "National salmon river" status (52 rivers) or where species of special
concern exist. Habitat simulation methods are commonly used for the case by case evaluations.
Recently, a holistic BBM method was applied in Norway (Alfredsen et al. 2011).
In France, the requirements to provide environmental flows in rivers are required by law using
the Tennant (1976) 10% of MAF rule. The rule is relaxed to 5% of MAF for rivers with
hydropeaking operations. In practice, project proponents are often mandated to carry out a
case study based on habitat simulation methods, using the EVHA model (Table1).
In the United Kingdom, two major projects have been carried out (i) to define water abstraction
limits that maintain a healthy river ecosystem and (ii) to define ecologically appropriate flow
releases from reservoirs (Acreman and Ferguson 2010). The limits for water abstraction were
developed for a number of river types (classification based on the requirements of fish,
macroinvertebrates and aquatic macrophytes), and different percentages of natural flow can be
abstracted in each river type. Moreover, the allowed abstraction limits vary between seasons,
and more water can be taken during higher flows than low flows (e.g. for river type "A1" in
winter, 35 % can be abstracted when flows >Q60, down to 20 % when flows < Q95; Acreman and
Ferguson 2010). Overall, allowable withdrawal ranges between 7.5% and 35 % of the natural
flow. For water releases from impoundments, a Building Block Methodology was adopted (see
4.2.3 Australia, South-Africa and New-Zealand
The environmental flows in Australia have strongly centered on holistic methodologies. Similarly
to Canada, Australia is divided into many jurisdictions who each describe environmental flows
based on separate criteria. All the jurisdictions have to provide environmental flows, and
although they have different legislation, every jurisdiction uses holistic methods and subscribes
to monitoring and adaptive management (A. Arthington, Griffith University, Australia, pers.
comm.). The common holistic methods used in Australia are reviewed in detail in Arthington
(1998). Environmental flows in South-Africa are also prescribed using holistic methods (e.g.
King and Brown 2006). In New Zealand, the environmental flow management is largely a
responsibility of regional councils (Snelder et al. 2011). The national environmental flow
standards are based on hydrological methods and the current set of guidelines was proposed in
2008. The New Zealand national guideline defines minimum flows and total allocation based on
proportions of the mean annual seven-day low flow (MALF). Environmental guidelines are
described separately for small (MAF < 5 m3/s) and large rivers (MAF > 5 m3/s), and are 90 %
and 30% of MALF (minimum flow / total allocation), and 80% and 50% of MALF for small and
large rivers, respectively (Snelder et al. 2011). The functionality of the general national guideline
has been criticized (Snelder et al. 2011).
A number of experts provided input to this document. The authors acknowledge the help from
the following people: T. Annear, A. Arthington, D. Boisclair, A. Bradford, M. Bradford, D. Caissie,
K. Clarke, K. Collet, K. Garroway, C. Gippel, J.H. Halleraker, P. Johnston, I. Jowett., E. Kampa,
H. Khan, M.S. Khan, J. King, T. Kondrat, M. Lebel, A. Locke, J. Long, C. MacNaughton, R.
Metcalfe, B. Raymond, K. Rollings, P. Parasiewicz, A. Paul, R. Ptolemy, K. Smokorowski, A. St-
Hilaire, P. Vezza, and R. Wysocki.
Acreman, M., and M.J. Dunbar. 2004. Defining environmental river flow requirements: a review.
Hydr. Earth Sys. Sci. 8: 861-876.
Acreman, M.C., and A.J. D. Ferguson. 2010. Environmental flows and the European Water
Framework Directive. Freshw. Biol. 55: 32-48.
Ahmadi-Nedushan, B., A. St. Hilaire, M. Berube, E. Robichaud, N. Thiemonge, and B. Bobee.
2006. A review of statistical methods for the evaluation of aquatic habitat suitability for
instream flow assessment. Riv. Res. Appl. 22: 503-523.
Ahmadi-Nedushan, B., A. St. Hilaire, M. Berube, T.B.M.J. Ouarda, and E. Robichaud. 2008.
Instream flow determination using a multiple input fuzzy-based rule system: A case study.
Riv. Res. Appl. 24: 279-292.
Alfredsen K., T.H. Bakken, and Å. Killingtveit, (editors). 1995. The River System Simulator.
User's Manual. SINTEF NHL report 1995.
Alfredsen, K., A. Harby, T. Linnansaari, O. Ugedal. 2011. Development of an inflow controlled
environmental flow regime for a Norwegian river. Riv. Res. Appl. Published online
Annear T.C., and A.L. Conder. 1984. Relative bias of several fisheries instream flow methods.
N. Am. J. Fish. Manage. 4: 531-539.
Annear T.C., W. Hubert, D. Simpkins, and L. Hebdon. 2002. Behavioural and physiological
response of trout to winter habitat in tailwaters in Wyoming, USA. Hydr. Proc. 16: 915-
Annear, T., I. Chisholm, H. Beecher, A. Locke, and 12 other authors. 2004. Instream flows for
riverine resource stewardship, revised edition. Instream Flow Council, Cheyenne, WY.
Anonymous. 2005. The science of instream flows: a review of the Texas instream flow program.
Committee on review of methods for establishing instream flows for Texas rivers, National
Research Council. 162 p.
Anonymous. 2007. La Salle River watershed. State of the watershed report. La Salle Redpoine
Conservation District. 296 p.
Apse, C., M. DePhilip, J. Zimmerman, and M.P. Smith. 2008. Developing instream flow criteria
to support ecologically sustainable water resource planning and management. Final
report to the Pennsylvania Instream Flow Technical Advisory Committee. 196 p.
Armanini, D.G., N. Horrigan, W.A. Monk, D.L. Peters, and D.J. Baird. 2011a. Development of a
benthic macroinvertebrate flow sensitivity index for Canadian rivers. Riv. Res. Appl. 27:
Armanini, D.G., W.A. Monk, D.E. Tenenbaum, D.L. Peters, and D.J. Baird. 2011b. Influence of
runoff regime type an a macroinvertebrate-based flow index in rivers of British Columbia
(Canada). Ecohydrology. Published online DOI: 10.1002/eco.234.
Armstrong, J.D. 2010. Variation in habitat quality for drift feeding Atlantic salmon and brown
trout in relation to local water velocity and river discharge. In: P. Kemp, and D. Roberts
(Editors). Salmonid Fisheries: Freshwater Habitat Management. Oxford: Wiley-Blackwell:
Armstrong, J.D., and K.H. Nislow. 2012. Modelling approaches for relating effects of change in
river flow to populations of Atlantic salmon and brown trout. Fish. Manage.Ecol. Published
online. DOI: 10.1111/j.1365-2400.2011.00835.x
Arthington, A.H. 1998. Comparative evaluation of environmental flow assessment techniques:
Review of holistic methodologies. LWRRDC Occasional Paper 26/98. Land and Water
Resources Research and Development Corporation (LWRRDC): Canberra. 46p.
Arthington, A.H. 2009. ELOHA down under: ecological limits of hydrological alteration by dams
in south-eastern Queensland, Australia. In Proceedings of the International Conference
on Implementing Environmental Water Allocations, 23-26 February, Port Elizabeth, South
Arthington, A.H., and B.J. Pusey. 2003. Flow restoration and protection in Australian rivers.
Riv. Res. Appl. 19: 377-395.
Arthington, A.H., J.L. Rall, M.J. Kennard, and B.J. Pusey. 2003. Environmental flow
requirements of fish in Lesotho Rivers using the DRIFT methodology. Riv. Res. Appl. 19:
Arthington, A.H., S.E. Bunn, N.L. Poff, and R.J. Naiman. 2006. The challenge of providing
environmental flow rules to sustain river ecosystems. Ecol. Appl. 16: 1311-1318.
Bain, M.B. 2007. Hydropower operations and environmental conservation: St. Marys River,
Ontario and Michigan. Project Report for the International Lake Superior Board of Control.
56pBain, M.B., and M.S. Meixler. 2008. A target fish community to guide river restoration.
Riv. Res. Appl. 24: 453-458.
Belzile, L., P. Bérubé, V.D. Hoang, and M. Leclerc. 1997. Méthode écohydrologique de
détermination des débits réservés pour la protection des habitats du poisson dans les
rivières du Québec. Joint Report INRS-Eau - Groupe conseil Génivar to Québec‘s Ministry
of Environment and to Dept. of Fisheries & Oceans, Canada. Report INRS-Eau 494, 82
pages +8 appendices.
Bérubé, P., M. Leclerc, and L. Belzile. 2002. Presentation of an ecohydrological method for
determining the conservative flow for fish habitats in Quebec‘s rivers (Canada). In:
Proceedings of the 4th IAHR International Symposium on Ecohydraulics. Capetown,
South Africa. 23p.
Blackburn, J., and P.M. Steffler. 2003. River2D: Two-Dimensional depth averaged model of
river hydrodynamics and fish habitat. Tutorial. University of Alberta. 40 p.
Booker, D.J., and M.C. Acreman. 2007. Generalisation of physical habitat-discharge
relationships. Hydrol. Earth Syst. Sci. 11: 141-157.
Borsanyi, P., A. Harby, O. Ugedal, and C. Craxner. 2004. A meso-scale habitat classification
method for production modelling of Atlantic salmon in Norway. Hydr. Appl. 14: 119–138.
Bovee, K.D. 1982. A guide to stream habitat analysis using the instream flow incremental
methodology. Instream Flow Information Paper 12, United States Fish and Wildlife
Service, Fort Collins, Colorado. 248 p.
Bovee, K.D., B.L. Lamb, J.M. Bartholow, C.B. Stalnaker, J.G. Taylor, and J. Henriksen.1998.
Stream habitat analysis using the instream flow incremental methodology: Biological
Resources Discipline Information and Technology Report USGS/BRD-1998-0004, VIII
Bradford, A. 2008. An ecological flow assessment framework providing a bridge to
implementation in Canada. Can. Wat. Res. J. 33: 215-232.
Bradford, M.J., and P.S. Higgins, J. Korman, and J. Sneep. 2011. Test of environmental flow
release in a British Columbia river: does more water mean more fish? Fresw. Biol. 56:
Brimley, B., J.-F. Cantin, D. Harvey, M. Kowalchuk, P. Marsh, T. Ouarda, B. Phinney, P. Pilon,
M. Renouf, B. Tassone, R. Wedel, and T. Yuzyk. 1999. Establishment of the Reference
Hydrometric Basin Network (RHBN) for Canada, Environment Canada.
Brisbane Declaration. 2007. Summary findings and a global action agenda of the 10th
International Riversymposium and International Environmental Flows Conference, 3-6
September 2007, Brisbane, Australia. Available at:
Caissie, D., and N. El-Jabi. 1995. Comparison and regionalization of hydrologically based
instream flow techniques in Atlantic Canada. Can. J. Civ. Eng. 22: 235-246.
Caissie, D., and N. El-Jabi. 2003. Instream flow assessment: From holistic approaches to
habitat modelling. Can. Wat. Res. J. 28: 173-183.
Caissie, D., N. El-Jabi, and C. Hebert. 2007. Comparison of hydrologically based instream flow
methods using a resampling technique. Can. J. Civ. Eng. 34: 66-74.
Castleberry, D.T., J.J. Cech Jr., D.C. Erman, D. Hankin, M. Healey, G.M. Kondolf, M. Mangel,
M. Mohr, P.B. Moyle, J. Nielsen, T.P. Speed, and J.G. Williams. 1996. Uncertainty and
instream flow standards. Fisheries 21: 20-21.
Clarke, K.D., T.C. Pratt, R.G. Randall, D.A. Scruton, and K.E. Smokorowski. 2008. Validation
of the flow management pathway: effects of altered flow on fish habitat and fishes
downstream from a hydropower dam. Can. Tech. Rep. Fish. Aquat. Sci. 2784. 97 p.
Clipperton, G.K., C.W. Koning, A.G.H. Locke, J.M. Mahoney, and B. Quazi. 2003. Instream flow
needs determinations for the South Saskatchewan River Basin, Alberta, Canada. Alberta
Environment, Edmonton, AB. 271 pp.
Cott, P.A., D.M.A. Monita, A.R. Majewski, B.W. Hanna, and K.J. Bourassa. 2005. Application of
the NWT winter water withdrawal protocol with bathymetric profiles of select small lakes in
the Mackenzie delta region. Can. Manuscr. Rep. Fish. Aquat. Sci. 2731: VII +73 p.
Conder A.L., and T.C. Annear. 1987. Test of weighted usable area estimates derived from a
PHABSIM model for instream flow studies on trout streams. N. Am. J. Fish. Manage.7:
Cunjak, R.A. 1996. Winter habitat of selected stream fishes and potential impacts from land-
use activity. Can. J. Fish. Aquat. Sci. 53 (Suppl. 1): 267-282.
de Kerckhove, D.T., K.E. Smokorowski, and R.G. Randall. 2008. A primer on fish habitat
models. Can. Tech. Rep. Fish. Aquat. Sci. 2817. 65 p.
DFO. 2010. DFO Protocol for winter water withdrawal from ice-covered waterbodies in the
Northwest Territories and Nunavut. 3 pp.
Dolloff, C.A., H.E. Jennings, and M. Owen. 1997. A comparison of basinwide and
representative reach habitat survey techniques in three southern Appalachian
watersheds. N. Am. J. Fish. Manage. 17: 339–347.
Dunbar, M.J., K. Alfredsen, and A. Harby. 2011. Hydraulic-habitat modelling for setting
environmental river flow needs for salmonids. Fish. Manage. Ecol. Published online DOI:
Eisner, A., C. Young, M. Schneider, and I. Kopecki. 2005. MesoCASiMiR – new mapping
method and comparison with other current approaches. In: A. Harby et al. (editors) COST
626 European Aquatic Modelling Network – Proceedings from the final meeting in
Silkeborg, Denmark, 19– 20 May 2005. Silkeborg: Natural Environment Research
Institute, 397 pp.
Elliott, C.M., and R.B. Jacobson. 2006. Geomorphic classification and assessment of channel
dynamics in the Missouri national recreational river, South Dakota and Nebraska: U.S.
Geological Survey Scientific Investigations Report 2006-5313, 66 p.
Enders, E.C., D.A. Scruton, and K.D. Clarke. 2009. The 'Natural Flow Paradigm' and Atlantic
salmon - Moving from concept to practice. Riv. Res. Appl. 25: 2-15.
Espegren, G.D. 1996. Development of instream flow recommendations in Colorado using
R2CROSS: Denver, CO, Water Conservation Board, 34 p.
Extence, C.A., D.M. Balbi, and R.P. Chadd. 1999. River flow indexing using British benthic
macroinvertebrates: a framework for setting hydroecological objectives. Regul. Riv. Res.
Manage. 15: 543-574.
Fausch, K.D. 1984. Profitable stream positions for salmonids: relating specific growth rate to
net energy gain. Can. J. Zool. 62: 441-451.
Fausch, K.D., J.R. Karr, and P.R. Yant. 1984. Regional application of an index of biotic integrity
based on stream fish communities. Trans. Am. Fish. Soc. 113: 39–55.
Fonstad, M.A., and W.A. Marcus. 2010. High-resolution, basin-extent observations of fluvial
forms and implications for process understanding. Earth Surf. Proc. Land. 35: 680-698.
Forseth, T., M. Stickler, O. Ugedal, H. Sundt, G. Bremset, T. Linnansaari, N.A. Hvidsten, A.
Harby, T. Bongard, and K. Alfredsen. 2009. Accidental close-down of the Trollheim
Hydropower Plant in July 2008. Effects on the fish populations in the River Surna. NINA
Report 435. 35p. [In Norwegian with english abstract]
Franzin, W.G. (Ed.). 2009. Estimating effects of water withdrawals from the lower Athabasca
River: Instream Flow Needs Technical Task Group Final Report. Laughing Water Arts and
Science, Inc. Report 06-2009. 202p.
Girard, V. 2009. Testing the performance of a statistical hydraulic model in Andean streams.
M.Sc. Thesis, Cranfield University, Cranfield, UK. 68 p.
Ginot, V. 1995. EVHA: Un logiciel d‘évaluation de l‘habitat du poisson sous Windows. Bull.
Fran. Pêc. Piscic. 339: 303-308.
Gippel, C.J. 2005. Environmental flows: managing hydrological environments. Encyclopedia of
Hydrological Sciences, p. 2953-2972. DOI: 10.1002/0470848944.hsa199
Gippel, C.J., and M.J. Stewardson. 1998. Use of wetted perimeter in defining minimum
environmental flows. Reg. Riv. Res. Manage. 14: 53-67.
Goater, L., C.W. Koning, A.G.H. Locke, J.M. Mahoney, and A.J. Paul. 2007. Aquatic
environment impact ratings: a method for evaluating SSRB flow scenarios—Red Deer
River case study. Alberta Environment and Alberta Sustainable Resource Development,
Alberta. 47 pp.
Guay, J.C., D. Boisclair, D. Rioux, M. Leclerc, M. Lapointe, and P. Legendre. 2000.
Development and validation of numerical habitat models for juveniles of Atlantic salmon
(Salmo salar). Can. J. Fish. Aquat. Sci. 57: 2065-2075.
Guensch, G.R., T.B. Hardy, and R.C. Addley. 2001. Examining feeding strategies and position
choice of drift-feeding salmonids using an individual-based, mechanistic foraging model.
Can. J. Fish. Aquat. Sci. 58: 446-457.
Haney, J. A., D.S. Turner, A.E. Springer, J.C. Stromberg, L.E. Stevens, P.A. Pearthree, and V.
Supplee. 2008. Ecological implications of Verde River flows. A report by the Verde River
Basin Partnership, Arizona Water Institute, and The Nature Conservancy: 124.
Hannah, D.M., B.P.G. Smith, A.M. Gurnell, and G.R. McGregor. 2000. An approach to
hydrograph classification. Hydr. Proc. 14: 317–338.
Harby, A., and H. Sundt. 2007. Habitat conditions for grayling and brown trout in lower Vojmån.
SINTEF Energy Research Technical Report TR A6588. [In Norwegian].
Harris, N.M., A.M. Gurnell, D.M. Hannah, and G.E. Petts. 2000. Classification of river regimes:
a context for hydroecology. Hydr. Proc. 14: 2831-2848.
Harvey, K.D., P.J. Pilon, T.R. Yuzyk. 1999. Canada‘s Reference Hydrometric Basin Network
(RHBN) in Partnerships in Water Resources Management, Nova Scotia: Canada.
Hatfield, T., A. Lewis, D. Ohlson, and M. Bradford. 2003. Development of instream flow
thresholds as guidelines for reviewing proposed water uses. Consultants report for British
Columbia Ministry of Sustainable Resource Management and British Columbia Ministry of
Water, Land and Air Protection. Victoria, BC. 88p.
Hayes, J.W., N.F. Hughes, and L.H. Kelly. 2007. Process-based modeling of invertebrate drift
transport, net energy intake, and reach carrying capacity for drifty-feeding salmonids.
Ecol. Model. 207: 171-178.
Heggenes J. 1990. Habitat utilization and preferences in juvenile Atlantic salmon (Salmo salar)
in streams. Reg. Riv. Res. Manage. 5: 341-354.
Heggenes J., Å. Brabrand, and S.J. Saltveit. 1991. Microhabitat use by brown trout, Salmo
trutta L. and Atlantic salmon, S. salar L., in a stream: a comparative study of underwater
and river bank observations. J. Fish Biol. 38: 259-266.
Henriksen, J.A., J. Heasley, J.G. Kennen, and S. Nieswand. 2006. Users' manual for the
hydroecological integrity assessment process software (including the New Jersey
assessment tools). U.S. Geological Survey Open-File Report 2006-1093: 71 p. Available
Hudson, H.R., A.E. Byrom, and W.L. Chadderton. 2003. A critique of IFIM—instream habitat
simulation in the New Zealand context. Science for Conservation 231. 69 p.
Hughes N.F., and L.M. Dill. 1990. Position choice by drift-feeding salmonids : Model and test for
Arctic grayling (Thymallus arcticus) in subarctic mountain streams, interior Alaska. Can. J.
Fish. Aquat. Sci. 47: 2039-2048.
Hvidsten, N.A. 1993. High winter discharge after regulation increases production of Atlantic
salmon (Salmo salar) smolts in the River Orkla, Norway. In: Production of juvenile Atlantic
salmon, Salmo salar, in natural waters. Can. Spec. Publ. Fish. Aquat. Sci 118: 175-177.
IPCC. 2007. The physical science basis. Fourth Assessment Report of the Intergovernmental
Panel on Climate Change (Eds S. Solomon, D. Qin, M. Manning, Z. Chen, M. Marquis,
K.B. Averyt, M. Tignor & H.L. Miller). Cambridge University Press, Cambridge.
Jenkins, A.R., and E.R. Keeley. 2010. Bioenergetic assessment of habitat quality for stream-
dwelling cutthroat trout (Oncorhynchus clarkii bouvieri) with implications for climate
change and nutrient supplementation. Can. J. Fish. Aquat. Sci. 67: 371-385.
Jorde, K. 1996. Ecological evaluation of instream flow regulations based on temporal and
spatial variability of bottom shear stress and hydraulic habitat quality. Proceedings of the
second International Symposium on Habitat Hydraulics, Ecohydraulics 2000, Quebec
City, June 1996, Volume B, pp. 163-174.
Jowett, I.G. 1989. River hydraulic and habitat simulation, RHYHABSIM computer manual. New
Zealand Fisheries Miscellaneous Report 49. Ministry of Agriculture and Fisheries.
Christchurch, New Zealand.
Jowett, I.G. 1997. Instream flow methods: a comparison of approaches. Reg. Riv. Res.
Manage. 13: 115-127.
Jowett, I.G., and M.J. Duncan. 2011.. Effectiveness of 1D and 2D hydraulic models for instream
habitat analysis in a braided river. Ecol. Eng. Published online DOI:
Jowett, I.G., D. Rowe, and D. West. 1995. Fishery flow requirements of the Tongariro River.
Consultancy report ELE301. National Institute of Water and Atmospheric Research,
Hamilton, New Zealand. 140p.
Kampa, E., J. von der Weppen, and T. Dworak. 2011. Water management, Water Framework
Directive & Hydropower. Issue Paper of the Common Implementation Strategy Workshop,
Brussels, Belgium, 13-14 September, 2011. 82 p.
Available at http://www.ecologic-events.de/hydropower2/background.htm
Katopodis, C. 2009. Status of flow management (or Instream Flow Needs) across the central
and Arctic region. Fisheries and Oceans Canada.
Keenleyside, M.H.A. 1962. Skin-diving observations of Atlantic salmon and brook trout in the
Miramichi river, New Brunswick. J. Fish. Res. B. Can. 19: 625-634.
Kendy, E. 2009. River types. ELOHA online Toolbox.
Available at http://conserveonline.org/workspaces/eloha/documents/river-types-0
Kendy, E., J.S. Sanderson, J.D. Olden, C.D. Apse, M.M. DePhilip, J.A. Haney, and J.K.H.
Zimmerman. 2009. Applications of the Ecological Limits of Hydrologic Alteration (ELOHA)
in the United States. In Proceedings of the International Conference on Implementing
Environmental Water Allocations, 23-26 February, Port Elizabeth, South Africa.
Kennard, M.J., S.J. Mackay, B.J. Pusey, J.D. Olden, and N. Marsh. 2010. Quantifying
uncertainty in estimation of hydrologic metrics for ecohydrological studies. Riv. Res. Appl.
Kilgour, B.W., J. Neary, D. Ming, and D. Beach. 2005. Preliminary investigations of the use and
status of Instream-Flow-Needs methods in Ontario with specific reference to application
with hydroelectric developments. Can. Man. Rep. Fish. Aquat. Sci. 2723, 30 p.
King, J., and C. Brown. 2006. Environmental flows: striking the balance between development
and resource protection. Ecology and Society 11: 26.
King, J.M., R.E. Tharme, and C.A. Brown. 1999. Definition and implementation of instream
flows. Thematic Report for the World Commission on Dams. Southern Waters Ecological
Research and Consulting: Cape Town, South Africa. 87p.
King, J., C. Brown, and H. Sabet. 2003. A scenario-based holistic approach to environmental
flow assessments for rivers. Riv. Res. Appl. 19: 619-639.
King, J. M., R.E. Tharme, and M.S. de Villiers, (Editors). 2008. Manual for the Building Block
Methodology (updated version). Water Research Commission Report No. TT 354/08.
Cape Town, South Africa.
Konrad, C.P., A.M.D. Brasher, and J.T. May. 2008. Assessing streamflow characteristics as
limiting factors on benthic invertebrate assemblages in streams across the western United
States. Freshw. Biol. 53: 1983-1998.
Lamouroux, N. 1998. Depth probability distributions in stream reaches. J. Hydr. Eng. 124: 224-
Lamouroux N., H. Capra. 2002. Simple predictions of instream habitat model outputs for target
fish populations. Freshw. Biol. 47: 1543-1556.
Lamouroux, N., and I.J. Jowett. 2005. Generalized instream habitat models. Can. J. Fish.
Aquat. Sci. 62: 7-14.
Lamouroux, N., Y. Souchon, and E. Herouin. 1995. Predicting velocity frequency distributions in
stream reaches. Wat. Resour. Res. 31: 2367-2375.
Lamouroux, N., S. Mérigoux, H. Capra, S. Dolédec, I.G. Jowett, and B. Statzner. 2010. The
generality of abundance‐environment relationships in microhabitats: a comment on
Lancaster and Downes (2009). Riv. Res. Appl. 26: 915‐920.
Lancaster, J., and B.J. Downes. 2010a. Linking the hydraulic world of individual organisms to
ecological processes: putting ecology into ecohydraulics. Riv. Res. Appl. 26: 385‐403.
Lancaster, J., and B.J. Downes. 2010b. Ecohydraulics needs to embrace ecology and sound
science, and to avoid mathematical artefacts. Riv. Res. Appl. 26: 921–929.
Leopold, L.B., and T. Maddock. 1953. The hydraulic geometry of stream channels and some
physiographic implications. United States Geological Survey pp. 1–57.
Lewis, A., T. Hatfield, B. Chillibeck, and C. Roberts. 2004. Assessment methods for aquatic
habitat and instream flow characteristics in support of applications to dam, divert, or
extract water from streams in British Columbia. Consultant‘s report prepared for the
British Columbia Ministry of Sustainable Resource Management and the British Columbia
Ministry of Water, Land and Air Protection.
Linnansaari, T. 2009. Effects of ice conditions on behaviour and population dynamics of Atlantic
salmon (Salmo salar L.) parr. Ph.D. Dissertation, University of New Brunswick,
Fredericton, N.B., Canada. 324 p.
Linnansaari, T., K. Alfredsen, M. Stickler, J.V. Arnekleiv, A. Harby, and R.A. Cunjak. 2009.
Does ice matter? Site fidelity and movements by Atlantic salmon (Salmo salar L.) parr
during winter in a substrate enhanced river reach. Riv. Res. Appl. 25: 773-787.
Linnansaari, T., A. Keskinen, A. Romakkaniemi, J. Erkinaro, and P. Orell. 2010. Deep habitats
are important for juvenile Atlantic salmon Salmo salar L. in large rivers. Ecol. Freshw. Fish
Locke, A., and A. Paul. 2011. A desk-top method for establishing environmental flows in Alberta
rivers and streams, Government of Alberta. iv + 94 p.
Maddock, I.P., M.A. Bickerton, R. Spence, and T. Pickering. 2001. Reallocation of
compensation releases to restore river flows and improve instream habitat availability in
the upper Derwent catchment, Derbyshire, UK. Reg. Riv. Res.Manage. 14: 417–441.
Maine DEP. (Maine Department of Environmental Protection). 2006. Chapter 587: In-stream
flows and lake and pond water levels.
Mäki-Petäys A., A. Huusko, J. Erkinaro, and T. Muotka. 2002. Transferability of habitat
suitability criteria of juvenile Atlantic salmon (Salmo salar). Can. J. Fish. Aquat. Sci. 59:
Mann, J.L. 2006. Instream Flow Methodologies: An Evaluation of the Tennant method for higher
gradient streams in the national forest system lands in the Western U.S. M.Sc. Thesis,
Colorado State University, Fort Collins, CO. 143 p.
Milhous, R.T. 1999. History, theory, use, and limitations of the physical habitat simulation
system. Proceedings of the Third International Symposium on Ecohydraulics.
Available at http://www.fort.usgs.gov/Products/Publications/4002/4002.pdf
Milhous, R.T., M.A. Updike, and D.M. Schneider. 1989. Physical Habitat Simulation System
Reference Manual—Version II. Instream Flow Information Paper 26. United States Fish
and Wildlife Service, Fort Collins, Colorado. 248 p.
Monk, W.A., P.J. Wood, D.M. Hannah and D.A. Wilson. 2007. Selection of river flow indices for
the assessment of hydroecological change. Riv. Res. Appl. 23: 113-122.
Monk, W.A., D.L. Peters, R.A. Curry, and D.J. Baird. 2011. Quantifying trends in indicator
hydroecological variables for regime-based groups of Canadian rivers. Hydr. Proc. 25:
Moyle, P.B., J.G. Williams, and J.D. Kiernan. 2011. Improving environmental flow methods used
in California. Federal Energy Regulatory Commission Relicensing. California Energy
Commission, PIER. CEC-500-2011-037.
Muotka T., A. Mäki-Petäys, P. Kreivi, and H. Högmander. 1999. Spatial associations between
lotic fish, macroinvertebrate prey and the stream habitat: a multi-scale approach. Bor.
Env. Res. 3: 371-380.
New Zealand Ministry for the Environment. 2008. Proposed national environmental standard on
ecological flows and water levels. New Zealand Ministry for the Environment Discussion
Document. X + 61 p.
Available at: http://www.mfe.govt.nz/publications/water/proposed-nes-ecological-flows-
Ohlson, D., G. Long, and T. Hatfield. 2010. Phase 2 framework committee report of the Lower
Athabasca River. 141p.
Olden, J.D., and N.L. Poff. 2003. Redundancy and the choice of hydrologic indices for
characterizing streamflow regimes. Riv. Res. Appl. 19: 101-121.
Olden, J.D., M.J. Kennard, and B.J. Pusey. 2011. A framework for hydrologic classification with
a review of methodologies and applications in ecohydrology. Ecohydrology. Published
online DOI: 10.1002/eco.251
Orth, D.J., and O.E. Maughan. 1981. Evaluation of the "Montana method" for recommending
instream flows in Oklahoma streams. Proc. Okla. Acad. Sci. 61: 62-66.
O'Shea, D.T. 1995. Estimating minimum instream flow requirements for Minnesota streams from
hydrologic and watershed characteristics. N. Am. J. Fish. Manage. 15: 569-578.
Paul, A.J. and A.G.H. Locke. 2009. Evaluation criteria for flow alterations in the lower Athabasca
River – Fish Habitat. Instream Flow Needs Technical Task Group Evaluation Criteria
Parasiewicz, P. 2001. MesoHABSIM: a concept for application of instream flow models in river
restoration planning. Fisheries 26: 6–13.
Parasiewicz, P. 2007. The MesoHABSIM model revisited. Riv. Res. Appl. 23: 893–903
Parasiewicz, P. 2008. Application of mesohabsim and target fish community approaches to
restoration of the Quinebaug River, Connecticut and Massachusetts, USA. Riv. Res. Appl.
Parasiewicz, P., and J.D. Walker. 2007. Comparing and testing results of three different micro
and meso river habitat models. Riv. Res. Appl. 23: 904–923.
Parker, G.W., D.S. Armstrong, and T.A. Richards. 2004., Comparison of methods for
determining streamflow requirements for aquatic habitat protection at selected sites on the
Assabet and Charles Rivers, Eastern Massachusetts, 2000–02. U.S. Geological Survey
Scientific Investigations Report 2004-5092. 72 p.
Peters, D.L., D.J. Baird, W.A. Monk, D.G. Armanini. 2012. Establishing standards and
assessment criteria for ecological instream flow needs in agricultural regions of Canada.
J. Environ. Qual. 41: 41-51
Pirhalla, D.E. 2004. Evaluating fish-habitat relationships for refining regional indexes of biotic
integrity: development of a tolerance index of habitat degradation for Maryland stream
fishes. Trans. Am. Fish. Soc. 133: 144–159.
Poff, N.L., and J.K.H. Zimmerman. 2010. Ecological responses to altered flow regimes: a
literature review to inform environmental flows science and management. Freshw. Biol.
Poff, N.L., J.D. Allan, M.B. Bain, J.R. Karr, K.L. Prestegaard, B.D. Richter, R.E. Sparks, and
J.C. Stromberg. 1997. The natural flow regime. Bioscience 47:769-784.
Poff, N.L., B.D. Richter, A.H. Arthington, S.E. Bunn, R.J. Naiman, E. Kendy, M. Acreman, C.
Apse, B.P. Bledsoe, M.C. Freeman, J. Henriksen, R.B. Jacobson, J.G. Kennen, D.M.
Merritt, J.H. O'Keeffe, J.D. Olden, K. Rogers, R.E. Tharme, and A. Warner. 2010. The
ecological limits of hydrologic alteration (ELOHA): a new framework for developing
regional environmental flow standards. Freshwat. Biol. 55:147-170.
Railsback, S.F., and J. Kadvany. 2008. Demonstration Flow Assessment: judgment and visual
observation in Instream Flow Studies. Fisheries 33: 217-227.
Reiser, D.W., T.A. Wesche, and C. Estes. 1989. Status of instream flow legislation and
practices in North America. Fisheries 14: 22-29.
Richter, B.D. 2010. Re-thinking environmental flows: From allocations and reserves to
sustainability boundaries. Riv. Res. Appl. 26: 1052-1063.
Richter, B.D., J.V. Baumgartner, J. Powell, and D.P. Braun. 1996. A method for assessing
hydrologic alteration within ecosystems. Conserv. Biol. 10: 1163-1174.
Richter, B.D., J.V. Baumgartner, R. Wigington, and D.P. Braun. 1997. How much water does a
river need? Freshw. Biol., 37: 231-249.
Richter, B.D., J.V. Baumgartner, D.P. Braun, and J. Powell. 1998. A spatial assessment of
hydrologic alteration within a river network. Reg. Riv. Res. Manage. 14: 329-340.
Richter, B.D., M.M. Davis, C. Apse, and C. Konrad. 2011. A presumptive standard for
environmental flow protection. Riv. Res. Appl. Published online. DOI 10.1002/rra.1511
Rollings, K.W. 2011. Hydrological low flow indices for the protection of fish and fish habitat in
Newfoundland and Labrador. Abstract submitted to the 64th Canadian Water Resources
Association National Conference June 27-30, 2011, St. John‘s NL, Canada. 11p.
Sanderson J.S., N. Rowan, T. Wilding, B.P. Bledsoe, W.J. Miller, and N.L. Poff. 2011. Getting
to scale with environmental flow assessment: the watershed flow evaluation tool. Riv.
Res. Appl. Published online. DOI 10.1002/rra.1542
Scruton, D.A., K.D. Clarke, L.M.N. Ollerhead, D. Perry, R.S. McKinley, K. Alfredsen, and A.
Harby. 2002. Use of telemetry in the development and application of biological criteria for
habitat hydraulic modeling. Hydrobiologia 483: 71-82.
Shang, S. 2008. A multiple criteria decision-making approach to estimate minimum
environmental flows based on wetted perimeter. Riv. Res. Appl. 24, 54-67.
Sheehan, K.R., and S.A. Welsh. 2009. An interpolation method for stream habitat
assessments. N. Am. J. Fish. Manage. 29: 1-9.
Snelder, T., D. Booker, and N. Lamouroux. 2011. A method to assess and define environmental
flow rules for large jurisdictional regions. J. Am. Wat. Res. Assoc. 47: 828-840.
Stalnaker, C.B., B.L. Lamb, J. Henriksen, K. Bovee, and J. Bartholow. 1995. The Instream Flow
Incremental Methodology: a primer for IFIM. Biological Report 29. United States National
Biological Service, Fort Collins, Colorado. 45 p.
Tennant, D.L. 1976. Instream flow regimens for fish, wildlife, recreation and related
environmental resources. Fisheries 1: 6-10.
Tessman, S.A. 1980. Environmental assessment, Technical Appendix E, in environmental use
sector reconnaissance elements of the Western Dakotas region of South Dakota study.
Water Resources Research Institute, South Dakota State University, Brookings, South
Tharme, R.E. 2003. A global perspective on environmental flow assessment: emerging trends
in the developments and applications of environmental flow methodologies for rivers. Riv.
Res. Appl. 19: 397-441.
Tharme, R.E., and J.M. King. 1998. Development of the Building Block Methodology for
instream flow assessments, and supporting research on the effects of different magnitude
flows on riverine ecosystems.Water Research Commission Report No. 576/1/98. 452 p.
Thompson, J.R., M.P. Taylor, K.A. Fryirs, and G.J. Brierley. 2001. A geomorphological
framework for river characterization and habitat assessment. Aquat. Conser. Mar.
Freshw. Eco. 11: 373-389.
Urabe, H., M. Nakajima, M. Torao, and T. Aoyama. 2010. Evaluation of habitat quality for
stream salmonids based on a bioenergetics model. Trans. Am. Fish. Soc. 139: 1665-
United States Geological Survey. 2005. Atlas of water resources in the Black Hills Area, South
Dakota; HA-747. 120p.
Available at: http://pubs.water.usgs.gov/ha747
USFWS. 1981. Interim regional policy for New England streams flow recommendations.
Memorandum from H.N. Larsen, Regional Director, Region 5, US Fish and Wildlife
Service (USFWS), Newton Corner, Mass.
Vadas R.L. Jr., and D.J. Orth. 2001. Formulation of the habitat suitability models for stream fish
guilds: Do the standard methods work? Trans. Am. Fish. Soc. 130: 217-235.
Vezza, P., P. Parasiewicz, M. Rosso and C. Comoglio. 2011. Defining minimum environmental
flows at regional scale: application of mesoscale habitat models and catchments
classification. River Res. Appl. Published online DOI: 10.1002/rra.1571
Webb, J.A., M.J. Stewardson, W.M. Koster. 2010. Detecting ecological responses to flow
variation using Bayesian hierarchical models. Freshw. Biol. 55: 108-126.
Wesche, T.A., and P.A. Rechard. 1980. A summary of Instream Flow Methods for fisheries and
related research needs. Water Resources Research Institute, University of Wyoming,
Laramie, Wyoming, 122 p.
Zafft, D.J., P. Braaten, K. Johnson, and T. Annear. 1995. Comprehensive study of the Green
River fishery between the New Fork River Confluence and Flaming Gorge Reservoir,
1991-1994. Completion Report. Wyoming Game and Fish Department, Cheyenne, WY.
Zhang, Y., A. H. Arthington, S. E. Bunn, S. Mackay, J. Xiab, and M. Kennard. 2011.
Classification of flow regimes for environmental flow assessment in regulated rivers: the
Huai River Basin, China. River Res. Applic. Published online DOI: 10.1002/rra.1483
Table 1. Commonly used environmental flow assessment methods, listed by a general category. A source
for each method is provided; * indicates that the reference is not the original source of the method, but
provides a comprehensive description to understand the basics of the method.
Aquatic Base Flow (ABF)
*Caissie and El-Jabi (1995)
*Caissie and El-Jabi (1995)
Belzile et al. (1997)
Median Monthly Flow (Q50)
*Caissie and El-Jabi (1995)
Range of Variability Approach (RVA)
Richter et al. (1997)
Sustainability Boundary Approach
and Presumptive Standard
Richter et al. (2011)
Wetted Perimeter Inflection Point Method
*Gippel and Stewardson (1998)
Flowing Perimeter Method
Gippel and Stewardson (1998)
PHABSIM (Physical HABitat SIMulation system)
RYHABSIM (River hYdraulic and HABitat SIMulation)
EVHA (EValuation de HAbitat)
RSS (River System Simulator)
Alfredsen et al. (1995)
CASIMIR (Computer Aided SIMulation
of habitat In Regulated streams)
Blackburn and Steffler (2003)
Eisner et al. (2005)
Generalized Habitat models (e.g. STATHAB)
Lamouroux and Jowett (2005)
Building Block Method (BBM)
Tharme and King (1998)
DRIFT (Downstream Response to Imposed Flow
King et al. (2003)
Arthington (1998; et al. 2006)
Poff et al. (2010)
Table 2. Comparison of four general categories of environmental flow assessment methodologies.
Examination of historic flow data to find
flow levels that naturally occur in a river
and can be considered "safe" thresholds
for flow abstraction
Whole rivers, applicable for
Examination of change in a hydraulic
variable, often wetted width, as a function
of discharge; the change in variable is
taken as a proxy for general quantity of
habitat in a river
Applied at a study site / river
segment scale, upscaling to
whole river level based on the
assumption of "representative"
sites. River specific.
Examination of change in the amount of
physical habitat for a selected set of target
species as a function of discharge
Applied at a study site / river
segment scale, upscaling to
whole river level based on the
assumption of "representative"
sites. River specific.
Examination of flows in an expert opinion
workshop leading to recommendation of
flows for all components of the riverine
ecosystem, including societal and
Whole rivers, applicable for
regional or river specific scales
$ - $$$$
Table 3. Flow recommendations as per the Tennant method. Based on Tennant (1976.)
Description of flows
Recommended flow regimen
(% of Mean Annual Flow)
Oct - Mar
Apr - Sept
Flushing or maximum
Fair or degrading
Poor or minimum
Table 4: Summary of Indicators of Hydrologic Alteration (IHA) variables and ecological influences (adapted from IHA help file)
IHA Parameter Group
1. Magnitude of monthly
Mean or median value for each calendar month
Habitat availability for aquatic organisms
Soil moisture availability for plants
Availability of water for terrestrial animals
Availability of food/cover for fur-bearing mammals
Reliability of water supplies for terrestrial animals
Access by predators to nesting sites
Influences water temperature, oxygen levels, photosynthesis in
2. Magnitude and duration
of annual extreme water
1-day, 3-day, 7-day, 30-day and 90-day mean
1-day, 3-day, 7-day, 30-day and 90-day mean
Number of zero-flow days
Base flow index: 7-day minimum flow/mean flow for
Balance of competitive, ruderal, and stress- tolerant organisms
Creation of sites for plant colonization
Structuring of aquatic ecosystems by abiotic vs. biotic factors
Structuring of river channel morphology and physical habitat
Soil moisture stress in plants
Dehydration in animals
Anaerobic stress in plants
Volume of nutrient exchanges between rivers and floodplains
Duration of stressful conditions such as low oxygen and
concentrated chemicals in aquatic environments
Distribution of plant communities in lakes, ponds, floodplains
Duration of high flows for waste disposal, aeration of spawning
beds in channel sediments
IHA Parameter Group
3. Timing of annual extreme
Julian date of each annual 1-day maximum and 1-
Compatibility with life cycles of organisms
Predictability/avoidability of stress for organisms
Access to special habitats during reproduction or to avoid
Spawning cues for migratory fish
Evolution of life history strategies, behavioral mechanisms
4. Frequency and duration
of high and low pulses
Number of low pulses within each water year
Mean or median duration of low pulses (days)
Number of high pulses within each water year
Mean or median duration of high pulses (days)
Frequency and magnitude of soil moisture stress for plants
Frequency and duration of anaerobic stress for plants
Availability of floodplain habitats for aquatic organisms
Nutrient and organic matter exchanges between river and
Soil mineral availability
Access for waterbirds to feeding, resting, reproduction sites
Influences bedload transport, channel sediment textures, and
duration of substrate disturbance (high pulses)
5. Rate and frequency of
water condition changes
Rise rates: Mean or median of all positive
differences between consecutive daily values
Fall rates: Mean or median of all negative
differences between consecutive daily values
Number of hydrologic reversals
Drought stress on plants (falling levels)
Entrapment of organisms on islands, floodplains (rising levels)
Desiccation stress on low-mobility streamedge (varial zone)
Table 5. The main steps identified in the Instream Flow Incremental Methodology (IFIM) framework
(Source: Hudson et al. 2003, based on Stalnaker et al. 1995).
Table 6. Examples of classification schemes used in mesohabitat modeling. A) Classification used in
the MesoHABSIM model (Parasiewicz 2007) and B) in the Norwegian mesohabitat model (Borsanyi et
al. 2004). The Norwegian classification system uses thresholds to differentiate the choices between
different categories (surface pattern: wave height above or below 5 cm; gradient: above or below 4 %;
velocity: above or below 0.5 m/s; depth: above or below 0.7 m).
Shallow stream reaches with moderate current velocity, some surface turbulence
and higher gradient. Convex streambed shape.
Higher gradient reaches with faster current velocity, coarser substrate, and more
surface turbulence. Convex streambed shape.
Stepped rapids with very small pools behind boulders and small waterfalls.
Moderately shallow stream channels with laminar flow, lacking pronounced turbulence.
Flat streambed shape.
Monotone stream channels with well determined thalweg. Streambed is longitudinally
flat and laterally concave shaped.
Uniform fast flowing stream channels.
Deep water impounded by a channel blockage or partial channel obstruction. Slow flow.
Concave streambed shape.
Where main flow passes over a complete channel obstruction and drops vertically to
scour the streambed.
Slack areas along channel margins, caused by eddies behind obstructions.
Channels around the islands, smaller than half river width, frequently at different
elevation than main channel.
Table 7. A list of initial points-of-contact in the different provinces of Canada with regard to environmental flows. * indicates that the contact does not
represent a province-specific association, but general environmental flow expertise.
Fisheries and Oceans Canada, Pacific Region
British Columbia Ministry of Environment, Aquatic
Conservation Science Section
Government of Alberta, Sustainable Resource Development,
Fish and Wildlife Division
Government of Alberta, Sustainable Resource Development,
Fish and Wildlife Division
Saskatchewan Watershed Authority
Manitoba Water Stewardship, Ecological Services Division,
Ontario Ministry of Natural Resources, Aquatic Research
and Development Section
University of Guelph
Fisheries and Oceans Canada, Central and Arctic Region
Ontario Ministry of Environment
M. S. Khan
Ontario Ministry of Environment
World Wildlife Fund
World Wildlife Fund
Canadian Rivers Institute / University of New Brunswick
Environment Canada / University of New Brunswick
Government of NB, Department of Natural Resources
Government of NB, Department of Environment
Fisheries and Oceans Canada, Gulf Region
Prince Edward Island
Government of PEI, Department of Environment, Labour and
Nova Scotia Environment, Water & Wastewater Branch
Newfoundland and Labrador
Government of NFL, Department of Environment and
Conservation, Water Resources Management Division
Newfoundland and Labrador
Fisheries and Oceans Canada, NL Region
Western Arctic Area
Fisheries and Oceans Canada, Western Arctic Area
Western Arctic Area
Fisheries and Oceans Canada, Western Arctic Area
Figure. 1. Cascading effects of flow alteration on physical and biological aspects of the river
ecosystem. In-water linkages affecting fish habitat are shown in boxes and endpoints of concern to fish
are in bubbles. Source: Clarke et al. (2008). [The reproduction is a copy of an official work that is
published by the Government of Canada and has not been produced in affiliation with, or with the
endorsement of the Government of Canada]
Figure 2. Conceptual relationships between flow and biological response as is assumed by different
environmental flow assessment methodologies. Historic flow refers to the duration curve exceedence
methods within the hydrological methodology group, and not to % MAF methods for which the
conceptual response between flow and biological response is assumed to be similar to the Hydraulic
methods (see Figure 3). Source: Jowett (1997). Copyright (2012): Wiley, reproduced with permission.
Figure 3. Tennant's (1976) relationship between proportion of mean annual flow and depth, velocity
and wetted width, based on 11 streams (38 different flows, 58 cross-sections) located in Montana,
Wyoming and Nebraska, USA. Copyright (2012): Taylor and Francis, reproduced with permission.
Figure 4. A) An illustration of the basic idea of Sustainability Boundary Approach (Source: Richter
2010, Copyright (2012): Wiley, reproduced with permission) and B) SBA applied as the Presumptive
Standard for environmental flow protection. Source: Richter et al. 2011, Copyright (2012): Wiley,
reproduced with permission.
Figure 5. Hypothetical wetted perimeter–discharge relationship plotted on two differently scaled sets of
axes. Different breakpoints are apparent, even though the same data are plotted on both graphs.
Source: Gippel and Stewardson (1998), Copyright (2012): Wiley, reproduced with permission.
Figure 6. A) Generalized approach used in habitat simulation methodology. (a) Hydro-geomorphology
of the study site(s) are surveyed and hydraulic models are calibrated, (b) Species-habitat association
models are developed for chosen variables, (c) weighted usable area is estimated as a function of
discharge for the select species and/or lifestages. Source: Girard (2009), reproduced with permission.
B) Habitat modelling approach using a fuzzy-logic based approach. The Habitat Suitability Indices are
replaced by expert derived fuzzy rules. Source: CASiMiR website http://www.casimir-
software.de/aufbau_eng.html, Copyright (2012): Schneider & Jorde Ecological Engineering,
reproduced with permission.
Figure 7. An example of the mesohabitat approach showing mapping results of different
hydromorphological units at four different discharges in a study site in Quinebaug River, Connecticut,
USA. Source: Parasiewicz (2008), Copyright (2012): Wiley, reproduced with permission.
Figure 8. A chart showing flow-habitat relationship derived using a mesohabitat modelling
(MesoHABSIM) approach. Source: Vezza et al. (2011), Copyright (2012): Wiley, reproduced with
Figure 9. Conceptual model of the Building Block Methodology approach. Using a natural hydrograph,
key components of the river flow are identified (A), and some proportion of these building blocks are
retained to provide an environmental flow regime. For example, blocks 1 and 6 are used to retain
perennial low flows of the hydrograph, blocks 2, 4 and 5 are used to separate wet and dry seasons,
and block 3 is used to introduce a first larger flood of the wet season. Source: King et al. (2008),
Copyright (2012): Water Research Commission, South-Africa, reproduced with permission.
Figure 10. A schematic of the four modules of the DRIFT framework. The example shows the fish
component of the Biophysical module; the specific sub-tasks within this module are different for each
biophysical discipline which all are contained within the module. Source: Arthington et al. (2003),
Copyright (2012): Wiley, reproduced with permission.
Figure 11. Conceptual model of the framework suggested by Arthington et al. (2006).Benchmark
conditions (vertical lines in graph (d)) can be identified for each flow variable using a number of
different river health indicators. See text for description of the different steps in the framework. Source:
Arthington et al. (2006), Copyright (2012): Ecological Society of America, reproduced with permission.
Figure 12. Conceptual figure of the ELOHA framework. See text for description of the different steps.
Source: Poff et al. (2010), Copyright (2012): Wiley, reproduced with permission.
Figure 13: Histogram displaying the number of stations against total number of years of record for river
gauging stations available through HYDAT.
Figure 14: (a) Map of existing and discontinued hydrometric gauging stations (rivers only); (b) map of
existing and discontinued RHBN stations.
Figure 15: (a) Map of six identified hydrological regime classes in Canada; (b) standardised weekly
runoff hydrographs for each of the six regime classes. Source: Monk et al. (2011), Copyright (2012):
Wiley, reproduced with permission.
Figure 16. Conceptual model of a 2-tiered environmental flow framework. The Level 1 assessment
serves as a coarse filter and identifies projects with "no HADD". Level 2 assessment requires case
specific analysis due to potential HADD. Source: Hatfield et al. (2003), reproduced with permission.
Appendix A: List of Abbreviations
ABF: Aquatic Base Flow
BBM: Building Block Methodology
DFO: (Department of) Fisheries and Oceans Canada
DRIFT: Downstream Response to Imposed Flow Transformation
EFC: Environmental Flow Components
ELOHA: Ecological Limits of Hydrologic Alteration
HADD: Harmful Alteration, Disruption or Destruction of fish habitat
HSI: Habitat Suitability Index
HYDAT: National hydrological database maintained by Water Survey of Canada
IFIM: Instream Flow Incremental Methodology
IHA: Indicators of Hydrologic Alteration
LAR: Lower Athabasca River
MAF: Mean Annual Flow
MMF: Mean Monthly Flow
PHABSIM: Physical HABitat SIMulation
POF: Percentage Of (natural) Flow
QX: Flow that is exceeded X% of the time
RHBN: Reference Hydrometric Basin Network
RVA: Range of Variability Approach
SBA: Sustainability Boundary Approach
WSC: Water Survey of Canada
WUA: Wetted Usable Area
Appendix B. The current method of establishing minimum or environmental flows in some European countries.
Table is originally published as the Annex II, p. 68-70 in Kampa et al. (2011). The attached table is a summary of a longer questionnaire directed to the
government officials in each country; the full response to questionnaire of each country is available at http://www.ecologic-
events.de/hydropower2/background.htm. Abbreviations of the various countries are used in the Table are: AT, Austria; BE, Belgium; BG, Bulgaria; CH,
Switzerland; CZ, Czech Republic; DE, Germany; ES, Spain; FI, Finland; FR, France; HU, Hungary; IS, Iceland; IT, Italy; LT, Lithuania; LU, Luxembourg; LV,
Latvia; NL, The Netherlands; NO, Norway; PL, Poland; PT, Portugal; RO, Romania; SE, Sweden; SI, Slovenia; SK, Slovakia; UK, United Kingdom.