Resource Directed Measures for Protection of Water Resources: River Ecosystems
R20: THE USE OF ECOLOGICAL INFORMATION ON FISH IN THE SPECIFICATION OF THE FLOW
COMPONENT OF THE RESERVE (DESKTOP, RAPID, INTERMEDIATE AND COMPREHENSIVE
DETERMINATIONS)
Senior Author:
C.J. Kleynhans, Institute for Water Quality Studies, Department of Water Affairs and Forestry
J.S. Engelbrecht, Mpumalanga Parks Board, Lydenburg
Editor:
Heather MacKay, Department of Water Affairs and Forestry
Lizette Guest, Guest Environmental Management
Version:
1.0
Date:
24 September 1999
M:\f_rdm_october\rivers\version 1.0\riv_appR20_version10.doc
Department of Water Affairs and Forestry, South Africa
Version 1.0: 24 September 1999
R20/1
Resource Directed Measures for Protection of Water Resources: River Ecosystems
Appendix R20:
The Use of Ecological Information on Fish in the Specification of
the Flow Component of the Reserve (Desktop, Rapid,
Intermediate and Comprehensive Determinations
The information provided here is mostly contained in the building block methodology manual (BBM: King and
Louw, in prep).
R20.1 Introduction
R20.1.1 The Role of Fish in River Structure, Functioning and Management
Because fish communities often include a range of species that represent a variety of trophic (feeding) levels
(i.e. omnivores, herbivores, insectivores, piscivores), they integrate the effect of detrimental environmental
changes. Their presence, therefore, can also infer the presence of other aquatic organisms since they occupy the
top of the food chain in most aquatic systems. They also pass through most trophic levels above the primaryproducer stage throughout their development from larvae to adults. Fish community structure can, therefore, be
regarded as reflective of integrated environmental health (Karr et al. 1986).
If the fish species richness and fish assemblage structure in a river or section of a river is respectively
sufficiently high and diverse, the requirements of fish can add considerably to the understanding of the
functioning of rivers and specification, monitoring and management of flow and flow related needs. In general,
it is also surmised that if management of the flow requirements of fish are successful, the requirements of
macro-invertebrates will also be satisfied due to the larger habitat scale requirements of fish.
It must be emphasized that the approach indicated in this chapter is mostly generic to make it possible to apply it
to different rivers across the country. In addition, quantitative ecological information on many of the indigenous
fish species of South Africa are limited. Consequently much use has to be made of expert knowledge and
professional opinion and it is highly recommended that the ecologist that do this assessment know the river and
its fish at least reasonably well.
R20.1.2 Role of Fish in the Determination of the Ecological Reserve
The general public as well as resource managers are usually more aware and concerned with the well-being and
existence of fish than with other forms of aquatic life. Consequently, aspects such as the ecological
requirements of fish as well as their use as indicators of ecological importance and sensitivity of a river and as
sentinels of ecological integrity, can be often be used to explain the reasoning behind the specification of
particular flows. In addition, as fish are good indicators of long-term (several years) effects and broad habitat
conditions due to their relative longevity and mobility, the monitoring of fish is often regarded as essential to
determining whether or not the goals of management have been achieved (Karr et al. 1986).
The role and use of fish as indicators of ecological flow requirements is the primary subject of this chapter.
However, fish as indicators of biological integrity and ecological importance and sensitivity are also touched on
here.
Ecological flow requirements and how this information can be used in terms of the specification of the
ecological reserve will be essentially be addressed according to the four levels of reserve determination, i.e. the
desktop and rapid estimates and the intermediate and comprehensive determinations.
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/2
Resource Directed Measures for Protection of Water Resources: River Ecosystems
R20.2 Ecological flow requirements of fish
The consideration of fish ecological information aims to link the size, temporal and spatial distribution of flow
or floods in a river to the ecological requirements of life-history stages of different fish species. The diversity
and usefulness of habitats required by the different fish species or life-history stages are usually a function of
base flow. For example, some fish species need permanent flow in a specific habitat during all their life stages
while others only need specific flow related events during migration, spawning or larval stages. On the other
hand, high flows and floods may be needed to initiate the development of gonads, facilitate migrations and
spawning and to activate and clean spawning beds and nursery areas. The following examples serve to illustrate
the role of fish in the estimation of flow requirements:

Fish species that are habitat and feeding specialists and/or are associated with riffles are likely to be most
dependent on permanent flow. This dependence can often be related to their water quality requirements
(i.e., oxygen concentration and water temperature). Some riffle species, such as the majority of Chiloglanis
spp., are dependant on permanent flow during all their life-history stages while other riffle species or certain
of their life stages can survive in pools during periods of low or even no-flow. These needs of fish give
important cues as to the size, distribution and constancy of base flow required during the low flow season.

Many fish species are dependent on sand/gravel beds or shallow slow flowing backwaters with inundated
marginal vegetation for breeding and nursery requirements. The temperature and volume of water in these
areas in the river are major factors contributing towards successful spawning and recruitment. Gonad
development and maturation followed by successful spawning may, however, also require a combination of
stimuli at different times such as increased current velocity, specific water quality changes (i.e., changes in
electrical conductivity, influx of organic substances from the catchment), maintenance of specific water
temperatures over a period of time, barometric pressure changes and releases of pheromones. However,
recruitment does not only depend on successful spawning but also on the maintenance of temperature and
flow over the period that is essential for the development of the embryos and larvae. The larval stages of
several fish normally favour nursery areas characterised by shallow, warm water with low velocities that are
rich with zooplankton. The timing and period that these stimuli and areas are available are important
considerations for setting base flows and floods during the breeding season.

Some estuarine fish species (i.e. glassies (Ambassis), pipefish (Microphis) and sleepers (Eleotris and
Hypseleotris)) need both the river and the estuary to complete their life cycles. These species have adapted
to breed during the low flow season to avoid or minimise unstable environmental conditions (i.e. floods). It
is surmised that it is crucial for the survival of these species that adequate flows are maintained during the
low flow months to maintain habitats for breeding and nurseries purposes in marginal vegetation.

Several truly catadromous fish species occur in several South African rivers. This includes the eels (i.e.
Anguilla mossambica, A. marmorata and A. bengalensis labiata) and the freshwater mullet (Myxus
capensis). It is of critical importance for the survival of these species that the adults are able to migrate
down to the sea to spawn and that the juveniles can migrate towards their freshwater feeding and maturation
areas in rivers. It is therefore extremely important that freshwater cues enter the ocean during this period to
attract the larvae of these fish.
It is surmised that by meeting the flow related habitat requirements of the above-mentioned sensitive species or
life-history stages in the river, the flow requirements of the majority of associated primary and secondary
freshwater fish species will also be satisfied.
R20.3 Fish ecological information and the levels of ecological
reserve determination
The four levels of ecological reserve determination as applicable to fish ecology are addressed here. However, at
the time of the compilation of this report, fish ecological information have only regularly and extensively been
used for the desktop estimate (i.e., quaternary catchment assessments) and the comprehensive reserve (i.e., for
the purposes of instream requirements for a number of rivers).
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/3
Resource Directed Measures for Protection of Water Resources: River Ecosystems
The usefulness of fish in terms of the ecological reserve can be viewed in the following context:

The conservation status of most fish species have been determined (Skelton 1987).

Their general ecological requirements are often relatively well known.

Their distribution patterns have been recorded for many rivers.

Assessment of biological integrity is often based on indices that makes use of attributes of fish assemblages
such as species richness, assemblage composition, trophic composition, habitat guilds and health and
condition. Of these the Index of Biological Integrity (IBI; Karr 1981) and its variations are commonly used
(Kleynhans 1999; Hughes & Oberdorff 1999). This kind of assessment, in combination with other groups of
biota, makes an essential contribution as to the understanding of the current ecological state of the river and
the formulation of an attainable ecological state or ecological management class.
Although four relatively distinct levels of reserve determination are indicated, in reality the distinction between
them may not always be clear. It is probably more realistic to view reserve determination with respect to fish as
a continuum with increasing information requirements and higher confidences (cf. Kleynhans 1999). The level
of reserve determination that will be used will depend on factors such as the current and desired ecological
integrity of the river and its ecological importance and sensitivity.
R20.3.1 Desktop Determination
The principle use of fish here is in the assessment of the ecological importance and sensitivity of a particular
river or part of a river. The presence of rare and endangered species, unique species and species (including
various life-history stages) with a particular sensitivity to flow (and flow related water quality aspects) is
considered in combination with other ecological information to provide an indication of the ecological
importance and sensitivity. Only available information is used for this purpose (i.e., published information and
professional knowledge and opinion) and no field surveys are undertaken. In cases where information for a
particular quaternary catchment (or other delineation) is lacking, information from similar catchments or river
sections that are better known, are used. Existing information on the integrity of fish assemblages are also used
in the assessment of the present ecological state of a river (cf. Kleynhans Appendix R11).
R20.3.2 Rapid Determination
In most respects the same considerations as for the desktop estimate will be valid here. However, the
requirement for information with a higher level of confidence will have to be met as the purpose for this
estimate will go beyond merely planning purposes. The best available existing information will be used but this
will usually have to be supplemented with limited fish surveys (i.e. at a limited number of sites and limited to
techniques and habitats amenable to rapid sampling and assessment) conducted specifically for this purpose. In
addition, it is envisaged that sampling will be limited to 1 -3 sites. As no hydraulic transects will be used for this
rapid estimate, it is important that fish habitat be assessed in as much detail as possible. Where possible flow
measurements should be made according to standard techniques (See Appendix R17) and photography and
videography be used to capture information on fish habitat.
R20.3.3 Intermediate Determination
The approach will be similar to the comprehensive reserve, but only a limited of sites will be sampled. Mostly,
only sites where hydraulic assessments will be made will be sampled and then only in a once-off fashion. It is
expected that 2 - 3 sites will be required for this determination. Sampling should aim to collect fish in at least a
qualitative but representative fashion at all available habitats at the selected sites. Photography and videography
should be used to capture information on fish habitat.
R20.3.4 Comprehensive Determination
Fish sampling and habitat assessment will be required at all sites used for hydraulic purposes. Usually about
four sites will be involved. However, it may be necessary to supplement these with sampling at sites with good
or excellent fish habitat but not suitable for hydraulic modeling. In addition, it may be necessary to sample fish
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/4
Resource Directed Measures for Protection of Water Resources: River Ecosystems
and/or assess habitat during different flow conditions so as to get a representative idea of conditions. Sampling
should aim to collect fish in a representative fashion at all available habitats at the selected sites. Photography
and videography should be used to capture information on fish habitat.
R20.4 The sequence of necessary activities
Activity 1: Assessment of Available Information






Determine the river or river sections of concern.
Determine what fish information is available for the river. Obtain information from Provincial Nature
Conservation organizations, museums, published accounts and local experts.
Identify the geomorphological zones in the river section of concern (see Appendix R2 and R18) and relate
the available fish distribution information to each of these zones.
Assess the distribution information and decide if it is representative of each zone and suitable for the
particular level of reserve determination.
Depending on the quality and detail of fish information, plan and conduct a fish survey (cf. Activity 2), then
collate all information (cf. Activity 3).
If fish information is sufficient (i.e., fish surveys have been recently carried out), collate all information (cf.
Activity 3). However, it is highly recommended that even when fish information is abundant and recently
collected, surveys still be conducted at the identified study sites (cf. Activity 3).
Activity 2: Plan and Conduct a Fish Survey
It must be emphasized that although surveys may be required to obtain particular information on the fish species
of a river, this is not a research programme per se.
Planning the survey

Based on the geomorphological zonation for the river, the study sites identified (i.e., sites used for hydraulic
analyses), topo-cadastral maps, the low altitude helicopter survey video and whatever historic information
available, tentatively identify sites that should be included in the survey. Sites that are reasonably
representative of habitat conditions in a zone should be aimed for. However, river sections known to be
impacted by antropogenic activities should be taken into account regarding the location of sites and if
possible, sites should be located upstream and downstream of such impacted sections. The study sites
should be regarded as the primary sites and other sites located according to the position of these sites.

It is not possible to prescribe the exact number of sites that should be included per river zone as this will be
influenced by factors such as the length of the river zones, the location of existing impacts and accessibility.
Expert knowledge and judgement should be relied on in these situations. However, a possible minimum
number of sites can include the study site per zone, as well as one site upstream and one downstream from
the study site in each zone.
Conducting the survey
The purpose of this section is to provide guidelines as to the kind of information required for reserve purposes
and is not an overview of sampling methods. Generally, fish surveys conducted for reserve purposes will also
not be aimed at providing a comprehensive assessment of the biological integrity of a river.

Except where seasonal information on flow requirements of fish species is required, it is strongly
recommended that wherever possible, fish surveys be conducted during the low flow season.

During the survey the habitats at sampling sites as identified in the previous section, should be assessed to
determine if it is representative of the zone and possible to sample. If this is not the case, a decision needs to be
taken as to selecting alternative sites.
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/5
Resource Directed Measures for Protection of Water Resources: River Ecosystems

At sampling sites the fish species caught (numbers and life-history stage) at each habitat with its associated
cover, should be recorded separately. Oswood and Barber's approach (1982) can be used to categorize flowdepth classes:
Slow (<0.3 m/s), shallow (<0.5 m): This includes shallow pools and backwaters.
Slow (<0.3 m/s), deep (>0.5 m): This includes deep pools and backwaters.
Fast (>0.3 m/s), shallow (<0.3 m): Shallow runs, rapids and riffles fall in this category.
Fast (>0.3 m/s), deep (>0.3 m): Deep runs, rapids and riffles fall under this category.
Cover can be categorized as follows:
Overhanging vegetation - thick vegetation overhanging water by approximately 0.3 m and not more than 0.1 m
above the water surface (Wang et al., 1996). Marginal vegetation is included here.
Undercut banks and root wads - banks overhanging water by approximately 0.3 m and not more than 0.1 m
above the water surface (Wang et al., 1996).
Stream substrate - various substrate components (rocks, boulders, cobbles, gravel, sand, fine sediment and
woody debris (“snags”)) that provide cover for fish.
Aquatic macrophytes - submerged and emergent water plants.
Activity 3: Collation and Analysis of Fish Distribution and Ecological Information
The ecological information required for this activity can be found in work such as that of Crass (1964), Jubb
(1967), Gaigher (1969), Pienaar (1978), Kleynhans (1984), Bell-Cross and Minshull (1988), Skelton (1993),
Russell (in prep) and Weeks et al. (1996). It is also advisable that the inputs of local experts be obtained in cases
where general literature information is not sufficient.
Flow and Habitat Requirements
The fish habitats found in each identified geomorphological zone should be described in general terms (i.e.
tabulated in terms of the presence and relative abundance of riffles, rapids, pools, backwaters, runs). Based on
the geomorphological zones the following approach should be followed for the collation of ecological
information on each of the fish species present in each zone:

Tabulate their habitat and flow requirements (riffles, rapids, pools, backwaters), including depth and flow
velocity requirements. Include available information on the habitat requirements of different life-history
stages (i.e. ova, larval, juvenile, subadult, adult). Provide quantitative specifications if possible, otherwise
follow a semi-quantitative or qualitative approach. Oswood and Barber's approach (1982) can be used to
categorize flow-depth classes (cf. Activity 2).

Tabulate cover requirements (substrate, marginal and overhanging vegetation, undercut banks, root wads,
etc.). Include available information on the habitat and cover requirements of different life-history stages.
Provide quantitative specifications if possible, otherwise follow a semi-quantitative or qualitative approach
(cf. Activity 2).

Identify fish species that are indicative of and can generally be considered representative and/or sensitive to
the availability of particular habitat conditions. If possible, identify species and life-history stages
representative of requirements for slow-shallow, slow-deep, fast-shallow and fast-deep flow-depth classes
and particular cover types.

Collate information on breeding requirements and characteristics of species where available (i.e. breeding
season, length of breeding season, fecundity, breeding stimuli, migration, spawning habitat,). The
association of any of these with particular flow and flood events should be indicated where possible.
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/6
Resource Directed Measures for Protection of Water Resources: River Ecosystems

Categorize and tabulate the tolerance of species towards flows based on the following:
Tolerant - species with no particular flow requirement during any life-history stage, i.e. they can survive
and reproduce in the absence of flow.
Moderately tolerant - species that require flow during particular life-history stages, i.e. for breeding and
migration.
Intolerant - species that have a requirement for flowing water during all life-history stages.

Categorize and tabulate the tolerance of species in respect of water quality changes based on the following:
Tolerant - species relatively hardy toward changes in water quality.
Moderately tolerant - species that can endure some changes in water quality. Some life-history stages may
be sensitive to water quality changes.
Intolerant - species that can endure only very limited changes in water quality. All life-history stages are
sensitive to water quality changes.

Categorize and tabulate the tolerance of species in respect of habitat and cover preferences based on the
following:
Tolerant - species with no particular habitat or cover preference.
Moderately tolerant - species with a preference for certain habitat and cover types during some life-history
stages.
Intolerant - species with specific habitat and cover requirements during all life-history stages.

Categorize and tabulate the tolerance of species in respect of trophic preferences based on the following:
Tolerant - species with no particular trophic preferences.
Moderately tolerant - species with moderately high trophic preferences.
Intolerant - species with high trophic preferences.
With this assessment it should be considered that a food resource may be in short supply in a particular segment,
require specialisation to be utilised and may be sensitive to environmental disturbance.
Assessment of Biological Integrity
The assessment of biological integrity using fish may not necessarily be as detailed as for biomonitoring surveys
as per the River Health Programme (Roux 1997). However, a collation of ecological information on the fish
assemblages in different sections of the river provides the opportunity to get a general approximation of the
biological integrity of a river. This information will be complementary to other biological assessment such as
those done for macro-invertebrates and the assessment of habitat integrity (See Appendix R5). For the
categorization of the biological integrity, the approach (fish assemblage integrity index, FAII) of Kleynhans
(1999) can be followed. However, such approaches will often only be possible where comprehensive sampling
can be done. Where this is not possible an alternative is proposed here. A qualitative assessment of the fish
assemblage information collected at the selected sites can be done based on the expected situation under
minimally impaired conditions, compared to the observed situation. This approach is similar to the habitat
integrity categorization (Table 1). The qualitative approach will usually be applicable to the rapid estimate of the
reserve. Specific requirements for the fish assemblage to be categorized into a particular category should be
formulated based on fish species present in a zone, at a site and in the habitats sampled at a site.
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/7
Resource Directed Measures for Protection of Water Resources: River Ecosystems
Table 1: Generic FAII assessment categories.
CATEGORY
DESCRIPTION OF GENERALLY EXPECTED CONDITIONS
A
Unmodified, or approximates natural conditions closely.
B
Largely natural with few modifications. A change in community characteristics may have
taken place but species richness and presence of intolerant species indicate little
modification.
C
Moderately modified. A lower than expected species richness and presence of most
intolerant species. Some impairment of health may be evident at the lower end of this
scale.
D
Largely modified. A clearly lower than expected species richness and absence or much
lowered presence of intolerant and moderately intolerant species. Impairment of health
may become more evident at the lower end of this class.
E
Seriously modified. A strikingly lower than expected species richness and general absence
of intolerant and moderately intolerant species. Impairment of health may become very
evident.
F
Critically modified. An extremely lowered species richness and an absence of intolerant
and moderately intolerant species. Only tolerant species may be present with a complete
loss of species at the lower end of the class. Impairment of health generally very evident.

The fish species expected to be present in a particular zone and habitats in a zone should be compiled and
used as the reference situation for minimally impaired conditions.

The fish species actually caught at a site and in a particular habitat type should be compared with the
expected situation.

Based on an analysis of the expected versus observed situation, the fish assemblage at the site should be
categorized according to the generic criteria in Table 1. Information as to the intolerance and habitat and
trophic preferences and requirements of species (cf. section on "Flow and habitats requirements") should be
taken into account in this categorization and the reasoning on which it is based, should be consistent and
explainable, even if it is not based on quantification.
Assessment of ecological importance and sensitivity based on fish
Conservation status and geographical distribution information on fish provide an opportunity to use this
knowledge in the assessment of the ecological importance of rivers. In addition, the available ecological
information on fish can be used in contributing to assessment of the sensitivity of rivers to various forms of
environmental disturbance. The following information should be collated for the purposes of ecological
importance and sensitivity assessments of the river and zones of the river being investigated:
The conservation status of fish species should be assessed based on the South African Red Data Book - Fish
(Skelton 1987). This information is directed at a National scale and assessment of the conservation status of
species on a regional level (Provincial and local) have usually not been determined formally and should be
obtained based on professional judgment and local knowledge.

The presence of unique species (i.e., isolated populations, endemics, genetically unique populations, etc.)
should be indicated. If possible this should be done on a National, Provincial and local scale.

The species richness (and abundances if the information is available) should be determined and compared
between the different zones, as well as with other rivers.

The information collated on the intolerance of fish species to various environmental conditions (cf. Flow
and habitat requirements) can be used to provide an indication of the sensitivity of the river to various forms
of disturbance. The presence of a large number of species with a requirement for flowing water during all
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/8
Resource Directed Measures for Protection of Water Resources: River Ecosystems
life-history stages are, for instance, an immediate general indication of the sensitivity of the river or zone to
flow modifications.
R20.5 Summary of what to include in the assessment
documentation
The purpose of the document is the creation of a perception of the ecological flow requirements of fish, and also
provide the data that can be used in the estimation of the ecological importance and sensitivity and biological
integrity for a particular river or section of river. This document should be concise and provide only the
information relevant to the investigation and should be based on tabulation of the collated information and short
discussions where appropriate.

List species expected to be present in the different zones, indicating their conservation status on a National,
Provincial and local scale, as well as indicators of uniqueness.

Habitat information for zones and each site sampled. In the case of sites sampled, indicate presence/absence
of flow-depth and cover classes. Where applicable, provide detailed information on study sites with
reference to hydraulic cross-sections.

Habitat (flow-depth and cover classes) preferences and requirements of fish species, also indicating
intolerances (cf. Flow and habitat requirements).

Species expected to be present and observed to be present at sampling sites. Categorize the biological
integrity (FAII) per site (and generalized per zone) according to Table 1. Where possible, provide detailed
information on study sites and relate fish requirements to hydraulic cross sections.

Provide conclusions on fish flow requirements (life-history stage requirements and habitat needs) and
sensitive and indicator species useful for this purpose (including different life-history stages).

Provide conclusions and explanations as to the current biological integrity class of the river as based on the
fish assemblages.
R20.6 Roles and responsibilities at the specialist meeting
In the case of the comprehensive reserve, at the specialist meeting a presentation of the current biological
integrity, ecological importance and the flow related requirements of fish in the various zones of the river should
be given. With the understanding that the participants have read the starter document, only the main points
should be indicated and the emphasis should be on the flow requirements of fish. However, where more detail is
required, such as during group discussions, the authors of the fish report should be capable and available to
explain and advise workshop participants in terms of flow or flood related habitat requirements of fish species
and life-history stages.
R20.7 Roles and responsibilities after the specialist workshop
The fish report is often compiled without having access to the work emanating from other work being done for
the specialist meeting (i.e. river geomorphology and classification, zonation and water quality assessment). In
addition, other fish ecology experts often form part of the specialist meeting and their insights and professional
knowledge may contribute significantly to the understanding and specification of the flow requirements of fish.
This information should be captured and incorporated into the final specialist meeting documentation. In some
cases it may be necessary to update the fish report based on information acquired during the meeting.
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/9
Resource Directed Measures for Protection of Water Resources: River Ecosystems
R20.8 Example terms of reference
(Not applicable to the Desktop Estimate)
Fish Report
Depending on the extent and aims of the investigation (i.e., the level of reserve determination), the terms of
reference for an assessment of the fish ecological attributes will be a report based on the following aspects:

A review of all geographical distribution records of fish in the river section of concern. If necessary (i.e.,
where information is insufficient), reference should also be made to related or similar stretches of
comparable rivers.

A survey of fish in all habitats at study sites. If study sites are not representative of all fish habitats,
additional sites should be sampled. This would include surveys in different geomorphological zones of the
river. Seasonal surveys may be required.

Indicate the conservation status of fish species based on a national, provincial and local scale.

Based on the current survey and historic information, provide an approximate estimate of the biological
integrity of the river and geomorphological zones of the river based on fish assemblages.

Based on available information and the current survey, collate as much flow related habitat requirements of
all species and life-history stages of all fish present in the river. Specific reference must be made to flow
related habitat requirements in terms of the hydraulic cross sections at study sites. Also included should be
specific spatial flow and flood requirements with reference to initiation of gonad development, facilitation
of migration and spawning and the maintenance and activation of spawning beds as well as nursery areas.
Specialist Meeting (Intermediate and Comprehensive Determination)

Explain and advise specialist meeting participants in terms of flow or flood related habitat requirements of
fish species and life-history stages.

Contribute towards the specification of the desired ecological management class with reference to fish
species.

Provide an assessment and explanation of the current biological integrity of the river as based on fish
assemblages.
Post Specialist Meeting (Intermediate and Comprehensive Determination)

Contribute towards the final meeting proceedings with regard to fish.

If necessary, modify the fish report based on new insights and information captured during the meeting.

Define the ecological endpoints for fish assemblages in river zones that will indicate if the ecological
management class is being attained based on the flow requirements of fish. These specifications will form
the basis of future monitoring programmes.
R20.9 Minimum and optimum specialist training

In cases where a fish survey is not required or only a very limited survey will suffice, it is essential that the
person responsible for the analysis and interpretation of data and report writing, be a person with
professional training and experience in fish ecology. Preferably this person should have experience in the
assessment and interpretation of information and data for ecological reserve specialist meeting purposes. If
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/10
Resource Directed Measures for Protection of Water Resources: River Ecosystems
this is not possible, the responsible person should have access to the advice of some-one who is experienced
in the approach with regard to fish.

Where a fish survey is required, it is necessary that such work be conducted by a trained and experienced
team under the supervision of the person responsible for the fish report. Ideally the ecologist responsible for
the fish report should be someone with professional training in this field, have been involved in previous
ecological reserve determination exercises and should also know the river and its fish species well.
R20.10
Potential pitfalls
Although fish can provide valuable information on ecological flow requirements, the following may present
problems:

The fish species richness in several of South Africa's rivers is very low. This is valid for rivers in several
geographical areas of South Africa and also for the upper reaches of most rivers in general. Consequently,
the amount of flow requirement information gained from such situations may be very limited and it may be
more realistic to place more dependence on the flow requirements of other biota which may be better
indicators of particular habitat conditions than the limited number of fish species present. In the same
context, many of South Africa's fish species are adapted to naturally diverse and environmentally harsh
conditions. As such they may not always be very good indicators of high flow sensitivity even if the species
richness is comparatively high.

Information on the ecological requirements of early life-history stages in particular of many South African
fish species are very limited. Often ecological deductions have to be made based on personal experience
with related species. There are inherent dangers and risks associated with using such a knowledge base and
it is advocated that a precautionary approach be followed in the specification of flow requirements of such
species.
R20.11
Further developments
Ecological flow requirement specification for South African fish species are predominantly based on expert
knowledge and experience. It is highly desirable that eventually these specifications be based on the
quantification of the ecological requirements of fish species. The ideal situation would be to have habitat
suitability curves (Bovee 1982) for all species and life-history stages. However, the ecological reserve
determination is not a research project and although a lot of information may be obtained during general fish
surveys and biomonitoring, specific research to provide such information and insights on the life-history
strategies are required.
R20.12
Monitoring
Monitoring of fish to determine if the desired ecological management class as specified and quantified
according to flows during inter alia the specialist workshop is attained, is actually an extension of the process
beyond the specialist meeting. However, one of the products of the meeting should be the specification of
ecological endpoints for fish assemblages in river zones that will indicate if the specified ecological
management class is being attained based on the flow requirements of fish. The development of a monitoring
programme should be based on these specifications and an expert system or decision support system should be
developed to assess the results of fish monitoring (Kleynhans 1999). Provision should be made in such a system
for feedback loops that provide the ability to refine flow specifications based on fish monitoring results as
information accumulate through time (i.e., thresholds of potential concern, cf. Rogers and Bestbier 1997).
Taking into account the scarcity of quantified ecological information on fish, development and refinement
through application and monitoring is seen as essential within the South African context.
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/11
Resource Directed Measures for Protection of Water Resources: River Ecosystems
R20.13
Conclusions
Fish can potentially provide a very important perspective on the flow requirements of rivers due the diversity of
life-history strategies that various species follow and the variety of habitats that they utilize during different lifehistory stages. Currently the use of fish in flow requirement specifications are often hampered by a lack of
quantified ecological information. Consequently the requirements of various life-history stages of many species
are not addressed in any detail. Species on which some ecological information is available are often used as
surrogates or indicators of general habitat requirements and conditions associated with it. To enable the better
utilization of fish in the ecological reserve determination process there is an urgent requirement for quantified
ecological information on most South African fish species.
References
Bell-Cross, G, & J.L Minshull. 1988. The fishes of Zimbabwe. Trustees of the National Museums and
Monuments of Zimbabwe, Harare, Zimbabwe.
Bovee, K.D. 1982. A guide to stream habitat analysis using the instream flow incremental methodology.
Instream Flow Information paper No. 21, US Fish and Wildlife Service, Biological Report 86, Washington, DC.
Crass, R.S. 1964. Freswater fishes of Natal. Shuter & Shooter, Pietermaritzburg.
Gaigher, I.G. 1969. Aspekte met betrekking tot die ekologie, geografie en taksonomie van varswatervisse in die
Limpopo- en Incomatiriviersisteem. Ph.D. thesis, Rand Afrikaans University, Johannesburg.
Hughes, R.M. & T. Oberdorff. 1999. Applications of IBI concepts and metrics to waters outside the United
States and Canada. Pp. 79 -93 In: T.P. Simon (ed.) "Assessing the sustainability and biological integrity of
water resources using fish communities". CRC press LLC.
Karr, J.R. 1981. Assessment of biotic integrity using fish communities. Fisheries 6:21-27.
Karr, J.R.., K.D. Fausch, P.L. Angermeier, P.R. Yant & I.J. Schlosser. 1986. Assessing biological integrity in
running waters: A method and its rationale. Illinois Natural History Survey, Special Publication.
Kleynhans, C.J. 1999. The development of a fish index to assess the biological integrity of South African rivers.
Water SA 25: 265-278.
Kleynhans, C.J. 1984. Die verspreiding en status van sekere seldsame vissoorte van die Transvaal en die
ekologie van sommige spesies. D.Sc. thesis, University of Pretoria, Pretoria.
Kleynhans, C.J. 1999. RDM Appendix R11, DWAF, IWQS, Pretoria.
Pienaar, U. de V. 1978. The freshwater fishes of the Kruger National Park. The National Parks Board of South
Africa.
Oswood, M.E. & W.E. Barber. 1982. Assessment of fish habitat in streams: goals, constraints, and a new
technique. Fisheries 7:8-11.
Rogers, K.H. and Bestbier, R. 1997. Development of a protocol for the definiation of the desired state of
riverine systems in South Africa. Department of Environmental Affairs and Tourism, Pretoria.
Roux, D.J. 1997. National aquatic ecosystem biomonitoring programme: overview of the design process and
guidelines for implementation. NAEBP Report series no. 6. Institute for Water Quality Studies, Department of
Water Affairs and Forestry, Pretoria, South Africa.
Russell, I.A. in preparation. Monitoring the conservation status and diversity of fish assemblages in the major
rivers of the Kruger Nationa Park. Ph.D. thesis, University of the Witwatersrand, Johannesburg.
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/12
Resource Directed Measures for Protection of Water Resources: River Ecosystems
Skelton, P.H. 1987. South African red data book - fishes. CSIR, SANSP, Report no. 137.
Skelton, P.H. 1993. A complete guide to the freshwater fishes of Southern Africa. Southern Book Publishers,
Halfway House, South Africa.
Wang, L., T.D. Simonson & J. Lyons. 1996. Accuracy and precision of selected stream habitat estimates. North
American Journal of Fisheries Management 16: 340-347.
Weeks, D.C., J.H. O'Keeffe, A.. Fourie & B.R. Davies. 1996. A pre-impoundment study of the Sabie-Sand
River System, Mpumalanga with special reference to predicted impacts on the Kruger National Park. Volume
One. The ecological status of the Sabie-Sand River System. Water Research Commission Report No. 294/1/96,
Pretoria.
Department of Water Affairs and Forestry, South Africa
Version 1.0 24 September 1999
R20/13