SIST EN 14757:2005

Water quality - Sampling of fish with multi-mesh gillnets

This European Standard describes a standardised method for sampling fish in lakes, using multi-mesh gillnets. The method provides a whole-lake estimate for species occurrence, quantitative relative fish abundance and biomass expressed as Catch Per Unit Effort (CPUE), and size structure of fish assemblages in temperate lakes. It also provides estimates, which are comparable over time within a lake and between lakes. This European Standard provides information on sampling routines, data handling and reporting, sampling of fish for age and growth analyses as well as applications and further treatment of data. Selected references in support of this European Standard are given in the Bibliography.

Wasserbeschaffenheit - Probenahme von Fisch mittels Multi-Maschen-Kiemennetzen

Diese Europäische Norm beschreibt ein genormtes Verfahren zur Probenahme von Fischen in Seen mit Multi-Maschen-Kiemennetzen. Dieses Verfahren liefert Daten zur Abschätzung des Artenspektrums im gesamten See, eine quantitative Abschätzung der relativen Artenhäufigkeit sowie der Biomasse, ausgedrückt als Einheitsfang (Catch Per Unit Effort, CPUE) und der Größenverteilung von Fischbeständen in Seen gemäßigter Zonen. Weiterhin liefert sie Abschätzungen, die für zeitliche Vergleiche innerhalb eines Sees und zwischen verschiedenen Seen geeignet sind. Diese Europäische Norm stellt Informationen für routinemäßige Probenahmen, den Umgang mit Daten und deren Protokollierung, die Probenahme von Fischen zur Alters- und Wachstumsbestimmung sowie Anwendungsmöglichkeiten und weitere Datenbehandlung zur Verfügung. Im Literaturverzeichnis sind ausgewählte Arbeiten angegeben, die direkten Bezug zur Anwendung dieser Europäischen Norm haben.

Qualité de l'eau - Echantillonnage des poissons a l'aide de filets maillants

La présente Norme européenne décrit une méthode normalisée pour l'échantillonnage des poissons dans les lacs, à l'aide de filets maillants multimailles. La méthode fournit une estimation, à l'échelle d'un lac, de l'occurrence des différentes espèces, de leur abondance relative, numérique et pondérale, exprimée en capture par unité d'effort (CPUE), et de la structure en taille des peuplements de poissons dans les lacs tempérés. Elle fournit également des estimations comparables dans le temps au sein d'un même lac ou entre plusieurs lacs. Elle donne aussi des informations concernant les routines d'échantillonnage, le traitement et la communication des données, l'échantillonnage des poissons en vue de l'analyse de l'âge et de la croissance, ainsi que les applications et le traitement ultérieur des données. La bibliographie présente une sélection d'ouvrages se rapportant à cette norme.

Kakovost vode – Vzorčenje rib s pomočjo zabodnih mrež (gillnet)

General Information

• Status: Withdrawn
• Publication Date: 30-Nov-2005
• Withdrawal Date: 09-Jul-2015
• ICS: 13.060.70 - Examination of biological properties of water
• ICS: 65.150 - Fishing and fish breeding
• Technical Committee: KAV - Water quality
• Current Stage: 9900 - Withdrawal (Adopted Project)
• Start Date: 09-Jul-2015
• Due Date: 01-Aug-2015
• Completion Date: 10-Jul-2015
• Directive: 2000/60/EC - Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy (substitutes Directive Reference 01/991 - Common Position on WFD)
• Directive: EUR-PUB - European Union - Public Standards
• Ref Project: EN 14757:2005 - Water quality - Sampling of fish with multi-mesh gillnets

Relations

• Replaced By: SIST EN 14757:2015 - Water quality - Sampling of fish with multi-mesh gillnets
• Effective Date: 01-Sep-2015

Standards Content (Sample)

SLOVENSKI STANDARD
01-december-2005
.DNRYRVWYRGH±9]RUþHQMHULEVSRPRþMR]DERGQLKPUHåJLOOQHW
Water quality - Sampling of fish with multi-mesh gillnets
Wasserbeschaffenheit - Probenahme von Fisch mittels Multi-Maschen-Kiemennetzen
Qualité de l'eau - Echantillonnage des poissons a l'aide de filets maillants
Ta slovenski standard je istoveten z: EN 14757:2005
ICS:
13.060.70 Preiskava bioloških lastnosti Examination of biological
vode properties of water
65.150 Ribolov in ribogojstvo Fishing and fish breeding
2003-01.Slovenski inštitut za standardizacijo. Razmnoževanje celote ali delov tega standarda ni dovoljeno.

EUROPEAN STANDARD
EN 14757
NORME EUROPÉENNE
EUROPÄISCHE NORM
August 2005
ICS 13.060.70; 65.150
English Version
Water quality - Sampling of fish with multi-mesh gillnets
Qualité de l'eau - Echantillonnage des poissons à l'aide de Wasserbeschaffenheit - Probenahme von Fisch mittels
filets maillants Multi-Maschen-Kiemennetzen
This European Standard was approved by CEN on 27 June 2005.
CEN members are bound to comply with the CEN/CENELEC Internal Regulations which stipulate the conditions for giving this European
Standard the status of a national standard without any alteration. Up-to-date lists and bibliographical references concerning such national
standards may be obtained on application to the Central Secretariat or to any CEN member.
This European Standard exists in three official versions (English, French, German). A version in any other language made by translation
under the responsibility of a CEN member into its own language and notified to the Central Secretariat has the same status as the official
versions.
CEN members are the national standards bodies of Austria, Belgium, Cyprus, Czech Republic, Denmark, Estonia, Finland, France,
Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania, Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Slovakia,
Slovenia, Spain, Sweden, Switzerland and United Kingdom.
EUROPEAN COMMITTEE FOR STANDARDIZATION
COMITÉ EUROPÉEN DE NORMALISATION
EUROPÄISCHES KOMITEE FÜR NORMUNG
Management Centre: rue de Stassart, 36  B-1050 Brussels
© 2005 CEN All rights of exploitation in any form and by any means reserved Ref. No. EN 14757:2005: E
worldwide for CEN national Members.

Contents page
Foreword .3
Introduction.4
1 Scope .5
2 Normative references .5
3 Terms and definitions.5
4 Principle.5
5 Sampling design and equipment .5
6 Time series sampling.7
7 Inventory sampling .10
8 Sampling routine.11
9 Data handling and reporting .12
10 Corrections for gillnet selectivity for six fish species.15
11 Estimate of sampling variance.17

12 Sampling fish for age- and growth analysis .18
13 Applications and further analyses.20
14 Limitations and supplementary sampling.21
Annex A (informative) Distribution of benthic multi-mesh gillnets at different depth strata in
lakes with different area and maximum depth.22
Annex B (informative) Example of forms for registration of fish and supplementary data .24
Bibliography.27

Foreword
This European Standard (EN 14757:2005) has been prepared by Technical Committee CEN/TC 230 “Water
analysis”, the secretariat of which is held by DIN.
This European Standard shall be given the status of a national standard, either by publication of an identical
text or by endorsement, at the latest by February 2006, and conflicting national standards shall be withdrawn
at the latest by February 2006.
According to the CEN/CENELEC Internal Regulations, the national standards organizations of the following
countries are bound to implement this European Standard: Austria, Belgium, Cyprus, Czech Republic,
Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Italy, Latvia, Lithuania,
Luxembourg, Malta, Netherlands, Norway, Poland, Portugal, Slovakia, Slovenia, Spain, Sweden, Switzerland
and United Kingdom.
Introduction
This is the second of several European Standards developed for evaluation of the species composition,
abundance and age structure of fish in rivers, lakes and transitional waters. Other standards describe
“Sampling of fish with electricity” (EN 14011) and “Guidance on the scope and selection of fish sampling
methods” (prEN 14962).
In most countries the use of the method specified in this European Standard requires permits from landowners
and national or regional authorities. In many countries permits are also required from authorities for animal
rights and animal welfare demands. Both fish diseases and diseases specific for other organisms, such as
freshwater crayfish, may be spread by placing equipment contaminated with pathogens or parasites in the
lake. The user of this method should check which national legislation is applicable.
WARNING — Persons using this European Standard should be familiar with normal laboratory and
fieldwork practice. This European Standard does not purport to address all of the safety problems, if
any, associated with its use. It is the responsibility of the user to establish appropriate safety and
health practices and to ensure compliance with any national regulatory conditions.
IMPORTANT — It is absolutely essential that tests conducted according to this standard be carried
out by suitably trained staff.
1 Scope
This European Standard describes a standardised method for sampling fish in lakes, using multi-mesh gillnets.
The method provides a whole-lake estimate for species occurrence, quantitative relative fish abundance and
biomass expressed as Catch Per Unit Effort (CPUE), and size structure of fish assemblages in temperate
lakes. It also provides estimates, which are comparable over time within a lake and between lakes. This
European Standard provides information on sampling routines, data handling and reporting, sampling of fish
for age and growth analyses as well as applications and further treatment of data. Selected references in
support of this European Standard are given in the Bibliography.
2 Normative references
The following referenced documents are indispensable for the application of this European Standard. For
dated references, only the edition cited applies. For undated references, the latest edition of the referenced
document (including any amendments) applies.
prEN 14962:2004, Water quality — Guidance on the scope and selection of fish sampling methods
3 Terms and definitions
For the purposes of this European Standard, the terms and definitions given in prEN 14962:2004 apply.
4 Principle
The sampling procedure is based on stratified random sampling. The sampled lake is divided in depth strata
and random sampling is performed within each depth stratum. Sampling of benthic fish is performed with
specially designed multi-mesh gillnets which are 30 m long and 1,5 m deep. The gillnets are composed of 12
different mesh-sizes ranging between 5 mm to 55 mm knot to knot following a geometric series. Gillnets used
for sampling pelagic fish are 27,5 m long and 6 m deep, with the smallest mesh-size being 6,25 mm. The
sampling effort needed to allow detection of 50 % changes in relative abundance between sampling occasions,
range between 8 gillnets per night (efforts) for small, shallow lakes, up to 64 efforts for lakes of about 5 000 ha.
When less accurate estimates of abundance are needed, an inventory sampling procedure may be used,
thereby reducing the number of efforts needed.
5 Sampling design and equipment
5.1 Sampling design
Fish are not randomly distributed over a lake. Depth distribution varies between fish species and with the
ontogeny of the fish. The horizontal distribution may also be influenced by habitat heterogeneity. Neither is the
distribution constant over the year, but will vary with temperature and season.
To cope with this uneven distribution a stratified random sampling design is used. The lake is stratified in
depth strata and a random sampling is performed within each depth stratum. Each gillnet is placed to
represent an independent sample of the fish assemblage. By randomising the location of each gillnet within
each depth stratum and by randomising the angle of the gillnet in relation to shoreline, an independent sample
of the fish in each stratum will be achieved. Randomisation is performed prior to fishing by the aid of depth
maps and a co-ordinate grid. If needed, the angle of the gillnet in relation to the shoreline shall be adjusted so
that the gillnet is within the corrected depth stratum.
5.2 Benthic gillnets
The multi-mesh gillnets have been designed for catching all types of freshwater fish species. Each gillnet is
composed of 12 different mesh-sizes ranging from 5 mm to 55 mm (knot to knot). The mesh-sizes follow a
geometric series, with a ratio between mesh-sizes of about 1,25. All gillnets have the same order of mesh
panels (see Table 1).
If experience has shown that large fish of certain species (e.g. bream, pike, tench) are difficult to catch with
the mesh sizes shown in Table 1, then these may be modified as required. However, a note on such
modification shall be given in the report (fishing protocol).
Table 1 — Mesh-size distribution (knot to knot) and thread diameter in the multi-mesh benthic gillnets
Mesh no Mesh size Thread diameter
mm mm
1 43 0,20
2 19,5 0,15
3 6,25 0,10
4 10 0,12
5 55 0,25
6 8 0,10
7 12,5 0,12
8 24 0,17
9 15,5 0,15
10 5 0,10
11 35 0,20
12 29 0,17
The gillnets shall be made out of homogeneous, uncoloured nylon. Each gillnet shall be 30 m long and 1,5 m
deep. Each mesh panel shall be 2,5 m long and mounted on a 30 m long buoyancy line (with a recommended
linear density in water of 6 g/m), and a 33 m long lead line (recommended linear density in air 22 g/m and in
water 9,9 g/m) made out of plastic in light grey colour. The diameter of the thread varies between 0,10 mm for
the 5 mm mesh, to 0,25 mm for the 55 mm mesh (Table 1). All mesh panels are commercially available. The
hanging ratio is 0,5 for all mesh sizes.
5.3 Pelagic gillnets
Each pelagic gillnet is 27,5 m long and 6 m deep. Gillnets used for sampling pelagic habitats are similar to the
benthic gillnets with the following exception. The smallest mesh (5 mm) has been excluded, because it has
not been possible to manufacture 5 mm panels mesh as deep as 6 m. The buoyancy line is 30 m, and the
lead line is 45 m with a hanging ratio of 0,5. The weight of the lines may be different from that of the benthic
nets. The pelagic nets are divided in half at 3 m depth by inserting a darkish colour (e.g. made of spun nylon)
in order to separate the catches below and above 3 m depth.
5.4 Time for sampling
The result of fish sampling using passive gears is highly influenced by water temperature, life history and time
for spawning of specific fish species. Therefore, the sampling period has to be chosen in such a way that each
single species is neither over nor under-represented in the catch. This means that the optimal sampling period
may differ between countries and regions. To minimise between-year variation, due to differences in activity
between species, the sampling period should be defined for each lake or region to be sampled in order to
make sampling data between different lakes and years comparable.
For instance, in northern Europe the sampling should normally take place between July 15 and August 31.
During this period most freshwater fish species living in lakes do not spawn, and the epilimnion temperature
usually exceeds 15 °C in most non-alpine areas. Due to decreasing epilimnion water temperature in
September it is not recommended to prolong the sampling period as the catch may decline substantially when
epilimnion temperature drops below 15 °C. Some species, and especially cyprinids, may also change their
behaviour during autumn, thereby affecting the representativity of the sampling. When it is known that the
catch is good for the species present even at temperatures down to 10 °C, then the sampling season may be
extended until September 15.
5.5 Sampling period
The setting time for the gillnets should ensure that the activity peaks of each fish species will be included. It
should also be so short that the fish does not degrade nor will be damaged by predatory fish while being
caught in the gillnet. This usually means that the gillnets should be set before dusk and lifted after dawn. To
avoid calculating abundance relative to hours of setting time, a standard fishing period of 12 h is
recommended. This is accomplished by setting the gillnets between 6 p.m. and 8 p.m. and lifting the nets
between 6 a.m. and 8 a.m.
In highly productive lakes with abundant fish populations, it may be necessary to shorten the setting time.
Otherwise the gillnets (or at least some mesh-panels in the gillnets) may be saturated with fish, thereby
affecting the outcome of the sampling. Saturation might bias the catch when more than 0,12 kg fish per m in
a 19 mm mesh, or 0,34 kg per m in a 70 mm mesh, are caught. Assuming a random distribution of fish over
all mesh-sizes, this means that saturation in a multi-mesh gillnet may start to affect the outcome when about a
6 kg fish is caught. In such cases, it is recommended to calculate the catch per unit effort (CPUE) relative to
hours of fishing time.
5.6 Gillnet selectivity
Correction factors for gillnet selectivity of the multi-mesh gillnets have been estimated for six fish species. The
sampling method provides abundance estimates only for fish larger than about 50 mm total length of fish
species catchable in gillnets. Abundance estimates of some less catchable species, such as eel (Anguilla
anguilla), burbot (Lota lota), bullhead (Cottus sp.) and pike (Esox lucius), as well as small Y-O-Y (young of the
year) individuals, may be underestimated.
6 Time series sampling
6.1 Sampling effort (gillnet-nights)
When the sampling aims at (1) quantifying relative abundance or biomass of different fish species, and (2)
comparing differences over time and between lakes, the variance of the estimate of the mean has to be
quantified. All fish should have the same probability of getting caught in a gillnet, and, therefore, a
representative sampling in a lake shall be performed. The number of gillnets used at each sampling occasion
is determined both by the minimum number of efforts needed to catch all catchable fish species and by the
required precision of the mean value. Usually the number of efforts needed to catch all catchable fish species
is lower than the number of efforts (net-nights) required to provide an acceptable precision of the estimate.
A commonly used minimum requirement for time series sampling has been to detect a 50 % difference
between sampling occasions in relative abundance of the most abundant fish. The amount of gillnet-nights
needed is determined by the precision, the lake area and the maximum depth of the lake. The higher the
desired precision is, and the larger and the deeper the lake is, the higher is the number of gillnet-nights. The
number of gillnets required to achieve a precision, which makes it possible statistically to determine a 50 %
difference between sampling occasions, is given in Table 2. By convenience the lakes have been divided into
six size classes (≤ 20, 21 ha to 50 ha, 51 ha to 100 ha, 101 ha to 250 ha, 251 ha to 1 000 ha, 1 001 ha to
5 000 ha), and the number of efforts based on multiples of 8, which is a normal workload for a one night
sampling made by two persons.
Table 2 — Number of efforts with benthic gillnets required to allow the detection of 50% changes
between sampling occasions in relation to lake area and maximum depth
Depth (m) Lake area (ha)
≤ 20 21 to 50 51 to 100 101 to 250 251 to 1 000 1 001 to 5 000
0 to 5,9 8 8 16 16 24 24
6 to 11,9 8 16 24 24 32 32
12 to 19,9 16 16 24 32 40 40
20 to 34,9 16 24 32 40 48 56
35 to 49,9 16 32 32 40 48 56
50 to 74,9  40 40 56 64
≥75   56 64
For small (<10 ha) and shallow lakes even 8 nets could overexploit the fish community, and especially deplete
the reproducing stock of certain species too much. The sampling effort should, however, never be less than 4
gillnets (see also 7.4).
Whole-lake estimates of the relative fish abundance in lakes larger than 5 000 ha usually require such a large
effort that it is practically impossible to use the recommended technique. In cases when larger lakes shall be
sampled, it is recommended that the lake is divided into separate basins, and that each basin is treated as a
separate lake. In large lakes, where whole-lake estimates of the fish fauna are not of main priority, sampling
can be performed at specific stations.
Stratification of gillnets is basically related to depth. The principles for depth stratification are given below. In
lakes with vegetation cover and in large shallow lakes, other stratification principles shall be considered.
However, it should be considered that depth is less variable over time than vegetation, and, therefore,
stratification related to vegetation shall be reconsidered at each successive sampling in a particular lake.
Reservoirs with steep banks may also be subjected to a modified stratification of gillnets.
6.2 Depth stratification of benthic gillnets
The depth zones are determined in relation to the volume of each stratum in such a way that each depth
stratum approximately equalises the same volume of water. Even if lake morphometry may vary considerably
between lakes, it is convenient to use a standardised scheme for stratification. For most lakes an
approximation of the depth strata can be based on morphometric lake data. Each lake is then divided in
approximately equal water volumes resulting in the following depth strata: 0 m to 2,9 m, 3 m to 5,9 m, 6 m to
11,9 m, 12 m to 19,9 m, 20 m to 34,9 m, 35 m to 49,9 m, 50 m to 74,9 m, > 75 m. Lakes deeper than 75 m are
rarely subjected to fish sampling using this type of benthic gillnets (see prEN 14962). The number of benthic
gillnets recommended in each depth stratum is given in Annex A. The table in Annex A includes optional
benthic gillnets at depth > 75 m in the largest lakes (251 m to 5 000 ha). Experience has shown that fish can
be caught in these nets, e.g. smelt, arctic char and bullhead. The information obtained from this effort should
be determined on a case-to-case basis.
To achieve a better estimate of the total fish abundance in lakes with extreme morphometry, the volume of
each depth stratum should be calculated, and the number of gillnets used at each stratum should be
distributed in relation to the volume of each stratum. Whenever the deepest stratum is too small to be used for
setting benthic gillnets which are independent of each other, it should be excluded in calculations of the total
number of gillnets used. When distributing gillnets over the lake, this depth stratum is treated as a part of the
stratum just above it.
6.3 Location of benthic gillnets
The location of each gillnet in the lake is determined in such way that the total catch should constitute an
unbiased sample of the catchable part of the fish assemblage in the lake. "Catchable" fish means fish species
which are usually caught in gillnets. Some predatory species with a typical ambush behaviour, such as
northern pike (Esox lucius), and some benthic species living very close to the bottom substrate, such as eel
(Anguilla anguilla), burbot (Lota lota) and bullhead (Cottus sp.), are often underrepresented in the gillnet catch.
Within the different depth strata, gillnets are set randomly over the whole lake. This could be performed by the
use of a pre-prepared co-ordinate grid placed over the depth map of the lake. By a randomisation procedure
each sampling location is located in each depth stratum, respectively (Figure 1). Gillnets are set in straight
lines and at random angles to the shoreline.
As the catch in each gillnet should be treated as an independent sample for that particular depth zone, no
gillnets shall be attached to each other.
6.4 Depth stratification of pelagic gillnets
To include samples also from the pelagic habitat, sampling with benthic gillnets should be supplemented by
sampling with pelagic gillnets in lakes with maximum depth greater than about 10 m. Even if there are no
apparent pelagic species in the lake, several fish species have a pelagic preference during part of their life
history. In contrast to sampling with benthic gillnets, the pelagic sampling does not provide an estimate over
the total water volume. Instead, pelagic sampling is performed as a depth profile over the deepest part of the
lake. The number of pelagic gillnets to be used is determined by the maximum depth of the lake. In more
shallow lakes, the benthic gillnets will provide a sufficient estimate of the pelagic fish in most cases.

Figure 1 — Morphometric map of a hypothetical 40 ha lake with 12 m maximum depth. Co-ordinate
grid, depth contours at 3 m, 6 m and 9 m, location of benthic gillnets (small marks) and pelagic gillnet
(large mark) are shown
7 Inventory sampling
7.1 General
The inventory sampling is a simplified method for fish sampling which will provide a rough estimate of the
occurrence and abundance of dominating fish species in the lake. This type of sampling may be used in
studies aimed at describing the distribution of species and in inventory studies, when the precision of the fish
abundance is of less importance.
7.2 Depth stratification
The depth stratification varies between species and may also vary between size classes within the same
species. Therefore, it is important that both the epi- and hypo-limnion in thermally stratified lakes are included
in the sampling effort, and that all depths of the lake are sampled, also when there is no clear thermal
stratification.
7.3 Location of gillnets
The benthic gillnets are distributed in the lake in such way that all types of habitats are sampled. Gillnets are
randomly set a) over the depth zone which covers the epi- and metalimnion, and b) in the hypo-limnion. Within
these two depth zones, the gillnets are set randomly over the whole lake. In the absence of a marked thermal
stratification, the same effort is used as if the lake has a meta-limnion. Each single gillnet is loosely set in a
straight line, at random angles from the shoreline.
The catch from each single gillnet shall comprise an independent sample. Therefore, the gillnets should not be
connected to each other.
7.4 Sampling effort
The number of efforts used is dependent on the number of gillnets needed to catch all catchable species in a
lake. Therefore, the lake area determines the size of the effort. In general, less than 4 gillnets should never be
used in a lake, independent of its size. However, in small unproductive small lakes even 4 gillnets may be
unacceptable to owners of the fishing rights. The lakes are divided into four size classes (< 50 ha, 51 ha to
300 ha, 301 ha to 2,000 ha, > 2,000 ha) to determine the sampling effort. In lakes larger than 5,000 ha an
inventory sampling has to be accomplished by other sampling methods. The lowest number of gillnet-nights
which should be used and the distribution of gillnets within the lake are given in Table 3. The effort may be
increased in order to increase the probability to catch all catchable fish species.
Table 3 — Minimum effort benthic gillnets (# of gillnet-nights) used in an inventory sampling in
relation to lake area
Lake area (ha) Number of gillnet-nights
Total In epi/metalimnion In hypolimnion
< 50 4 2 2
51 to 300 8 4 4
301 to 2 000 16 8 8
> 2000 24 12 12
8 Sampling routine
8.1 Pre-sampling
In order to maximise the output of the sampling effort a thorough planning shall precede all fish sampling.
When a lake has been selected for sampling, permission from the fishing right owner(s) has to be obtained. If
the responsible persons are informed about the aim and magnitude of the fishing activities, and the results are
communicated to responsible persons afterwards, this should normally not be an obstacle. To mitigate the
spreading of diseases due to fishing activities, a risk assessment for dispersion of pathogens has to be made.
Both fish diseases and diseases specific for other organisms, such as freshwater crayfish, may be spread by
placing equipment contaminated with pathogens or parasites in the lake.
If there is already a map of the lake with depth contours, this could be used to determine the total number of
efforts needed, and to determine if pelagic gillnets should be used. The map with depth contours is used to
divide the lake into appropriate depth strata and to determine the number of efforts that should be used at
each stratum. If the lake is being sampled for the first time, a randomisation of the gillnet locations should be
performed in advance. If the lake has been sampled earlier, the locations of the gillnets should as much as
possible resemble the earlier distribution in the lake. If data on depth of the lake is lacking, the sampling has to
be preceded by a sounding. This could be performed using a simple echo sounder and by running the boat in
predetermined transects over the lake before gillnets are set for the first time.
Supplementary information about the lake and the surroundings should be collected before sampling if
possible. All types of geographical and water chemical information should be collected. Especially information
about the fishery in the lake and on introduced fish species should be collected.
8.2 Sampling
All gillnets should be set between 6 p.m. to 8 p.m. Benthic gillnets are set randomly relative to the shore line at
the predetermined locations, and the depth of the most shallow and deepest points of the net are recorded
(Figure 1). The distribution of gillnets on each fishing night should be such that all depth strata are included, in
order to avoid bias due to differences in weather conditions between nights. A GPS instrument is
recommended to locate and record gillnet positions. Pelagic gillnets are set over the deepest part of the lake.
During the first night, gillnets are placed at depth 0 m to 6 m. The second night they are lowered to 6 m to
12 m and so on until the whole water column has been sampled according to Figure 2. Usually it is possible
for two experienced fishermen to fish with eight benthic gillnets and two pelagic gillnets per night in oligo- to
mesotrophic lakes. In eutrophic, highly productive lakes, the number of efforts per night has to be reduced
because the catch is usually so large that it will not be possible to rinse the gillnets and handle the fish within
the day after. In very deep lakes this method may require to long a time, and alternative sampling designs
may be used (see prEN 14962).
The day after setting, the gillnets are lifted at 6 a.m. to 8 a.m. After landing the nets, they are rinsed and the
fish are collected separately in marked net bags for each gillnet. If the fish are going to be used for gillnet
selectivity studies, the fish have also to be kept separated by mesh size. After the nets have been rinsed they
should be cleaned and dried until the next setting. Further treatments of the fish are performed as soon as
possible. If the weather is warm, the caught fish have to be kept cold, either in a cold-storage room or by use
of ice. After fish have been recorded and processed for further examinations, the gillnets are set again
between 6 p.m. and 8 p.m.
Figure 2 — Schematic view of the setting of pelagic multi-mesh gillnets, 6 m deep and 27,5 m long.
The gillnets are set over the deepest part of the lake, and lowered by 6 m on each consecutive day of
fishing
During all fish sampling the safety instructions for fieldwork on sea should be followed. There should always
be at least two persons able two swim on board the fishing vessel. The personnel should be equipped with life
jackets, a device for communication such as a mobile phone or a flag, megaphone or whistle to alert people
on land, and a first-aid box.
9 Data handling and reporting
9.1 Fish data
For each sampling occasion the number of the gillnet used, the geographical localisation of each gillnet in the
lake, and the maximum and minimum depth for each single gillnet are recorded (for forms, see Annex B). The
localisations of gillnets are also marked on a lake map with depth contours or as co-ordinates if GPS (Global
Positioning System) equipment is used.
The catch within each gillnet is registered as number of individuals and total weight for each species.
Optionally the catch within each mesh panel is registered in such a way that it is possible to track each
specific individual back to the gillnet and specific mesh panel in which it was caught. This will be of importance
if a more detailed correction for gillnet selectivity is to be performed. Accordingly, the total length for each
single specimen is registered in such way that each individual could be tracked back to the individual gillnets
(and if desirable also mesh panel) in which it was caught. Optionally, the wet weight of each specimen could
be recorded in a similar way. Total lengths are determined to the nearest mm and weight to the nearest gram.
Raw data should not be processed before they are stored in a database. Data should preferably be stored in a
database using lake (or lake ID), date for fishing, and gillnet number as ID-variables. Using the lake map, it
will then be possible to describe the exact location in the lake where the specific individual was caught.
Table 4 — Minimum requirements for fish data registration and reporting
List of fish species caught
A list of species caught in the gillnets should always be provided. As the sampling technique is based on a
passive system, the probability to getting caught varies among species and the species list may therefore not be
used as a definite list of fish species in the lake. However, the effort (number of gillnetnights) is calculated so
that on average all catchable species are caught at least on one occasion, which make the list comparable
between years.
Total number of caught fish:
The total number of each species.
Total weight of caught fish
The total weight of each species.
Number per Unit Effort (NPUE)
The simplest way to calculate NPUE is the arithmetic mean for the catch of each species. The variance
estimates will be larger if stratification is considered. By estimating mean and variance for each single depth
stratum, the variance may be minimised (see 9.1). NPUE should also be given as the number of the fish caught
in each depth strata in a way that it is possible to calculate the mean value for the lake and to describe depth
distribution of each species.
Weight per Unit Effort (WPUE)
Should be calculated similarly as for NPUE.
Length (and/or weight) frequency distributions:
Length (and/or weight) frequency distributions should be given for all dominant species in the lake. When there
is a special interest for some species, the frequency distributions could be corrected for gillnet selectivity (see
9.1). However, usually the difference between corrected length distributions and non-corrected distributions is of
minor importance for many species when the general fish population structure should be given.

9.2 Supplementary data
The outcome of the fish sampling is affected by physical/geographical factors such as lake size and depth,
water transparency, temperature and weather conditions during sampling. Therefore, supplementary data
should always be provided. This is shown in Table 5. Secchi disc depth and a temperature profile should be
recorded at each sampling occasion. A current weather report for the sampling occasion, including strength
and direction of the wind should also be registered.
Table 5 — Supplementary data used in assessment of fish sampling data
Geographical information
Lake identification
Name and number of the lake (co-ordinates in national grid system or longitude-latitude).
Watershed identification
Name and number of water system (drainage area code)
Altitude
Altitude is given in m above sea level. Preferably data from national geographical or hydrological institutes
are used
Lake area
The area of the lake should be given according to accepted references. If the area substantially deviates
from the area measured from maps or by other sources, both areas and references should be given.
Lake depth
If available both maximum and average depth should be given in m. If no published data are available, data
obtained during fish sampling using e.g. echo sounder may be given as preliminary data.
Physical data (usually measured once during fish sampling)
Vegetation
Coverage and plant infested volume shall be given when appropriate
Water transparency (at deepest part)
Water transparency, usually is measured as Secchi disc depth, given in fractions of a m.
Temperature (at deepest part)
A temperature profile is registered at 0,5 and then at each full m - 1 m, 2 m, and so on down to 25 m depth.
Water chemistry (at deepest part)
When available, water chemistry data should be added to the fish sampling. Water quality data reflecting
nutrient load (phosphorous and nitrogen), oxygen depletion (oxygen at hypolimnion) and acidification status
(pH, alkalinity and/or ANC, Acid Neutralisation Capacity) are preferable.
Sampling information
Date for fishing
First and last date for setting and lifting the gillnets should be given. By convenience the first sampling date
may be used as ID-variable in the database.
Number of efforts
The total number of gillnet-nights (efforts) used at different depth strata in the sampling should be recorded.
Often the standardised scheme is violated, and in order to determine the size of the error it is important to
include data on gillnet distribution.
Type of gillnets used
If standardised gillnets are used, the length, weight and depth of the gillnets are known. This makes it
possible to calculate the catch in terms of caught fish per m . If other types of multi-mesh gillnets are used
this information has to be added.
Type of sampling design
The type of sampling design (Time series/ Inventory sampling) should be given, as it is part of the quality
control. If neither of the two designs is followed it should be marked as "unclassified".
Time for gillnet setting
The time for setting and lifting the gillnets in the lakes should be given with an hourly precision. This makes it
possible to calculate the catch in relation to hour instead of "night".
Responsibility
The performer and institute responsible for the sampling should always be given.

A map with depth contours showing the location and unit efforts number of each gillnet should be added to
each sampling occasion (see Figure 1). The quality of the map should be such that the sampling could be
repeated without additional knowledge.
9.3 Databases and quality control
Data from the fish sampling should be stored in specially designed databases. A quality control should always
accomplish data storing, thereby minimising typing errors and avoiding preposterous data. The purpose is to
provide data of high quality for international, national, local and regional investigations and reports.
It is recommended that all activities in the fish sampling procedure be subjected to a quality assurance
programme in order to produce consistent results of high quality. The quality control should include all parts of
the sampling including training of fishermen, handling of equipment, fieldwork, handling of fish, analyses, data
handling, and reporting.
10 Corrections for gillnet selectivity for six fish species
10.1 Selectivity of multi-mesh gillnets
When using a passive sampling gear, the outcome of the sampling will be dependent upon the movements of
the fish and the mechanical properties of the gear to catch and to retain the fish. The properties of the gear
will affect the composition of the sample, and only a particular part of the population will be collected in the
sample. This means that the sampled population may not be the same as the biological population of interest.
Selectivity of gillnets include any process that causes the probability of being sampled to vary with the
characteristics of a fish. For a passive gear, selectivity usually is divided into a) encounter probability, b) the
probability of being caught in the mesh, and c) the probability of being retained in the gillnet after being caught.
Selectivity of the multi-mesh gillnets has been estimated for several fish species. It may be expected that the
condition of the fish may affect the gillnet selectivity due to changes of shape of the fish. However, differences
in condition (i.e. fish shape) between lakes have no practical effect on the catch composition because these
gillnets are composed by mesh-sizes following a geometric series. Adjacent mesh sizes would then cover
each other and correct for this error.
10.2 Corrections for gillnet selectivity for six fish species
Corrections for gillnet selectivity have been estimated for six fish species; two percid species (European perch,
Perca fluviatilis) and ruffe (Gymnocephalus cernuus), one cyprinid species (roach, Rutilus rutilus) and three
salmonid species; brown trout (Salmo trutta), Arctic char (Salvelinus alpinus) and smelt (Osmerus eperlanus).
The correction factors reflect to some extent the differences in body shape between the different species. For
example the spinier ruffe has a steep selectivity curve, whereas the more slender roach and the salmonid
species have more flat selectivity, but increasing, curves (Table 6). Perch has the flattest curve of all species.
Table 6 — Pooled Relative Efficiency (PRE) curves for multi-mesh gillnets for six freshwater fish
species estimated by fitting a 3rd order polynomial equation. Length (L) measured in mm
Species Function Range
mm
2 3
E. perch (Perca fluviatilis) PRE=0,4167+0,00128*L+0,00093*L -1,53E-05*L 40 to 380
2 3
Ruffe (Gymnocephalus cernuus) PRE=0,02862 +0,2735*L-0,03936*L +0,00179*L 40 to 140
2 3
Roach (Rutilus rutilus) PRE=0,2599+0,021*L-0,00121*L +3,76E-05*L 50 to 330
2 3
Brown trout (Salmo trutta) PRE=0,6449+0,05121*L-0,03936*L -5,52E-05*L 80 to 340
2 3
Arctic char (Salvelinus alpinus) PRE=0,4077+0,00351*L+0,000658*L +3,96E-06*L 60 to 300
2 3
Smelt (Osmerus eperlanus) PRE=-1,6278+0,3031*L+0,00428*L -3,70E-04*L 90 to 170

These relations can be used for reconstructing a more probable size distribution of each single species.
Because selectivity curves have not been estimated for all fish species, corrections are usually made only in
cases when a specific species is assessed.
rd
In practice, the gillnet selectivity is corrected by a species-specific 3 order polynomial function. The
correction of CPUE is calculated by using the polynomial estimate of relative length frequency distribution
(RLFD) for each species given in Table 7. Corrected values of CPUEs are achieved by multiplying the
observed number of specimens by the relative efficiency (RLFD) within each length class. The resulting
product is rounded to the nearest integer and pooled together over the length classes. The corrected biomass
for each length class is calculated by multiplying the corrected frequency with the mean individual biomass of
that length class. An example of the measured and corrected length distribution of a theoretical perch
population is shown in Figure 3. Depending on the slope of the correction factor, the small sized fish has a
larger correction factor compared to the large sized fish.
Table 7 — Gillnet selectivity for six common fish species caught in multi-mesh gillnets. Relative
Length Frequency Distribution (RLFD) curves estimated by fitting a third order polynomial equation.
Length (L) is measured in mm
Species Function Range
mm
2 3
E. perch (Perca fluviatilis) RLFD=1,7159-0,04595*L+0,00031*L -4,82E-06*L 40 to 380
2 3
Ruffe (Gymnocephalus cernuus) RLFD=1,36386-0,10525*L+0,01897*L -1,13E-03*L 40 to 140
2 3
Roach (Rutilus rutilus) RLFD=1,5285-0,01547*L-0,00074*L +7,96E-06*L 50 to 330
2 3
Brown trout (Salmo trutta) RLFD=1,25629+0,04187*L-0,00440*L +7,18E-05*L 90 to 330
2 3
Arctic char (Salvelinus alpinus) RLFD=1,48571-5,32E-05*L-0,00220*L +3,98E-05*L 60 to 300
2 3
Smelt (Osmerus eperlanus) RLFD=1,02857-3,69E-05*L-0,00153*L +2,76E-05*L 90 to 170

Key
X Length of perch in cm
Y Number of individuals
1 Corrected
2 Measured
Figure 3 - Example of measured and corrected length distribution of a theoretical perch population.
The correction increased the estimated total number of caught fish by 16 %

The observed means of CPUE in both number and biomass are generally smaller than the c
...