Aquaculture Reports 35 (2024) 101996
Available online 22 February 2024
2352-5134/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license ( http://creativecommons.org/licenses/by-
nc-nd/4.0/ ).
Condition factor tailored to lumpfish ( Cyclopterus lumpus ) used as cleaner
fish in salmonid farms
Solveig Engebretsen
a
,
*
, Magne Aldrin
a
, Liss Lunde
b
, Marthe Austad
c
, Trond Rafoss
d
,
Ole Roald Danielsen
e
, Andreas Lindhom
e
, Lauris Boissonnot
b
, Peder A. Jansen
a
,
f
a
SAMBA, Gaustadalleen 23A, Oslo 0373, Norway
b
Aqua Kompetanse AS, Storlavika 7, Flatanger 7770, Norway
c
VAL FoU, Hestvikvegen 73, Kolvereid 7970, Norway
d
Department of Natural Sciences, Post box 422, Kristiansand 4604, Norway
e
Norsk Oppdrettsservice AS, Abelnes 60, Flekkefjord 4404, Norway
f
Aqualife R & D, Havnegata 9, Trondheim 7010, Norway
ARTICLE INFO
Keywords:
Lumpfish
Cleaner fish
Condition
Fulton ’ s K
ABSTRACT
Lumpfish ( Cyclopterus lumpus ) are extensively used as part of the control measures against salmon lice in fish
farms. In recent years, there has been an increased focus on lumpfish welfare, and how to increase survival of
lumpfish during production. To survey lumpfish welfare and compare welfare between different fish farms,
welfare indicators are necessary. Condition factor is an attractive welfare indicator since it depends only on the
length and weight of the fish, which is easy to measure and does not require euthanisation. Fulton ’ s K is
commonly used to assess body condition for fish. However, this condition factor assumes isometric growth,
which has been found not to be a reasonable assumption for lumpfish. In this study, we suggest an alternative
expression for lumpfish body condition, which is based on almost 30 000 lumpfish sampled from Norwegian fish
farms during production. The resulting condition factor is given as 5.85 ⋅ W ∕ L
2.5016
, where W is the lumpfish
weight measured in grams, and L is the standard length measured in centimetres. We demonstrate that our
proposed condition factor is more suitable for comparing body condition of lumpfish than Fulton ’ s K, since
Fulton ’ s K is negatively correlated to length such that small fish tend to have high Fulton ’ s K factors. We
conclude that Fulton ’ s K is an inappropriate indicator of lumpfish welfare, and propose to rather use a condition
factor tailored to lumpfish, like the presently proposed condition factor. We also illustrate why standard length
(length up to the caudal fin) is more appropriate for measuring body condition than total length (i.e. length
including the caudal fin). For condition based on standard length, we found that caudal fin erosion was less
prevalent among the lumpfish with comparatively higher body condition. The same relationship was not found
for body condition based on total length.
1. Introduction
Lumpfish are commonly used as a control measure against salmon
lice ( Powell et al., 2018; Brooker et al., 2018; Barrett et al., 2020 ).
Satisfactory lumpfish welfare in fish farms is essential for justifying their
use. Lumpfish mortalities and infectious diseases pose a threat to the use
of cleaner fish in the fish farming industry ( Scholz et al., 2018; Klakegg
et al., 2020 ). Hence, there has been an increased interest in studies
focusing on lumpfish welfare in recent years ( Garcia de Leaniz et al.,
2022; Eliasen et al., 2020; Imsland et al., 2018b , 2019 , 2018a , 2020;
Klakegg et al., 2020 ).
Lumpfish is considered the most appropriate cleaner fish species for
salmonids farmed at low temperatures ( Powell et al., 2018; Imsland
et al., 2016; Brooker et al., 2018 ). Lumpfish is the most commonly used
cleaner fish species in Norway, and constituted 60% of the 46 million
cleaner fish deployed in Norwegian salmonid farms in 2021 ( Fisker-
idirektoratet, 2022 ). Note, however, that there has been a slight
decrease in lumpfish use the most recent years (i.e. lumpfish constituted
70% of a total of 61 million cleaner fish deployed in Norwegian farms in
2019 ( Fiskeridirektoratet, 2022 )). Another advantage with lumpfish is
that they are relatively easy to farm in hatcheries, as they grow relatively
fast ( Powell et al., 2018 ): almost all the lumpfish used in Norwegian
* Corresponding author.
E-mail address: [email protected] (S. Engebretsen).
Contents lists available at ScienceDirect
Aquaculture Reports
journal homepage: www.elsevier.com/loc ate/aqrep
https://doi.org/10.1016/j.aqrep.2024.101996
Received 11 October 2023; Received in revised form 3 February 2024; Accepted 18 February 2024
Aquaculture Reports 35 (2024) 101996
2
salmonid farms are farmed ( Fiskeridirektoratet, 2022 ).
In order to improve lumpfish welfare, there is a need for quantitative,
practical, and easy-to-use welfare measures ( Garcia de Leaniz et al.,
2022 ). An indicator of fish welfare is the condition factor, which is a
function of the weight and the length of the fish. It is used to assess the
fatness of the fish, such that given the length of the fish, the heavier fish
are assumed to be in better body condition ( Blackwell et al., 2000;
Bolger and Connolly, 1989; Le Cren, 1951 ). Condition factors have been
found to correlate with body composition, energy, and fat content
( Blackwell et al., 2000 ). Condition factors have for example been used to
compare populations under different environments, as an indicator of
changes in nutrition under different feeding conditions ( Bolger and
Connolly, 1989 ). Body condition can thus be used as an indicator to
compare and monitor welfare of lumpfish in different fish farms, for
example to assess whether there is enough available food. In a recent
survey of 31 lumpfish farmers, a majority reported use of condition
factor as a welfare indicator ( Garcia de Leaniz et al., 2022 ).
One of the most commonly used condition factors in aquaculture is
Fulton ’ s condition factor, hereby denoted by Fulton ’ s K ( Bolger and
Connolly, 1989; Ricker, 1975 ) (see Nash et al. (2006) for the history of
Fulton ’ s K). There are also several examples of studies applying Fulton ’ s
K to compare lumpfish body condition ( Imsland et al., 2022, 2014;
Flores-García et al., 2022; Geitung et al., 2020 , 2020 , 2018a ). Fulton ’ s K,
however, assumes that the shape of the fish does not change with length
(isometric growth), which in general is not a reasonable assumption for
most fish species ( Le Cren, 1951 ). Specifically, Fulton ’ s K has been found
not to be appropriate for lumpfish due to their allometric growth and
shape ( Daborn and Gregory, 1983 ).
Gutierrez-Rabadan et al. (2021) analyse the length-weight relation-
ship for lumpfish, in order to assess the deviation between lumpfish
weight and expected weight based on length. This was used to classify
lumpfish as emaciated. From their reported results, it is possible to
derive an expression for the condition factor, even though this was not
done in the study. In the present study, we are interested in the lumpfish
body condition post-deployment in sea, for which they analyse 355
lumpfish. We will denote the condition factor derived from their re-
ported results as Gutierrez-Rabadans ’ s K.
Guiterrez-Rabadan ’ s K has recently been recommended as an oper-
ational welfare indicator, particularly in Norway ( Boissonnot et al.,
2023 , 2022a ). Based on the 355 lumpfish, Gutierrez-Rabadan et al.
(2021) provide cut-offs for when lumpfish can be considered as severely
underweight or emaciated. Boissonnot et al. (2023 , 2022a) use these
cut-offs to generate a welfare indicator for condition, along with adding
a cut-off for an additional level of classification. As a sanity check, the
cut-offs were also assessed by visual inspection of emaciation ( Bois-
sonnot et al., 2022a ). The expression for Guiterrez-Rabadan ’ s K used in
Boissonnot et al. (2023 , 2022a) is
Guiterrez - Rabadan
′
s K = 9 . 057 W ∕ L 2 . 559 ,
where L is the total length measured in centimetres (i.e. length including
the caudal fin) of the lumpfish and W is the weight measured in grams.
This has been used to classify condition as a score between 0 and 3,
where 0 is good condition ( K ≥ 1), 1 is slightly emaciated K ∈ [0.9, 1), 2 is
clearly emaciated K ∈ [0.75, 0.9) and 3 is severely emaciated ( K < 0.75)
( Boissonnot et al., 2023 , 2022a ). However, Boissonnot et al. (2023)
point out that this expression for the condition factor is subject to large
uncertainty, and hence analysing a larger data set could be used to
decrease the uncertainty.
Total length is often used to measure lumpfish length. However,
when length is used to calculate body condition, including the caudal fin
in the length measurement may have unfortunate effects due to poten-
tial caudal fin erosion. Imagine two lumpfish with the same weight and
the same standard length (i.e. length up to but not including the caudal
fin), but where one has no damage on the caudal fin and the other has a
completely eroded caudal fin. Then the lumpfish with the eroded caudal
fin will have a higher measured body condition than the lumpfish with
the caudal fin intact, if total length is used to compute body condition.
We study this by analysing a data set where weight, standard length,
total length and caudal fin erosion were measured for 93 randomly
sampled lumpfish from a commercial salmon farm in Norway. We also
use these data to provide a factor for converting between standard and
total length of lumpfish.
The aims of our study are four-fold. By analysing a data set of almost
30 000 lumpfish in Norwegian salmonid farms, we establish an
expression for the condition factor which is tailored to lumpfish. In this
data set, length is measured by standard length. This is an updated
version of the data set used in Engebretsen et al. (2023) to analyse the
relationship between salmon lice grazing of lumpfish and different
operational conditions. Based on the 30 000 observations, we can then
obtain a distribution for the condition factors of lumpfish deployed in
Norwegian fish farms, allowing assessment of how a specific lumpfish
compares to the population of lumpfish deployed in Norwegian fish
farms in condition. Secondly, we demonstrate why the commonly used
Fulton ’ s K is not appropriate for lumpfish. Thirdly, we assess whether
the length-weight relationship found in Gutierrez-Rabadan et al. (2021)
holds for a much larger, independent data set of lumpfish. Finally, we
investigate the relationship between caudal fin erosion and body con-
dition measured by both standard and total length.
2. Data
2.1. Data set for establishing a lumpfish condition factor
The data were collected in an online digital form. More details on the
data collection are provided in Engebretsen et al. (2023) . The data set
used in this analysis consists of 29 669 observations of lumpfish length
and weight sampled between April 6th, 2020 and January 26th, 2023
from 57 different Norwegian salmonid farms.
For the 29 669 observations of lumpfish, we have information on
weight measured in g, and length measured in cm. As we want to
establish an expression for the condition factor of lumpfish, it is neces-
sary to have a standard definition of lumpfish length. Lumpfish length
was here measured by the standard length, that is, length was defined as
the length only to the caudal peduncle, not including the caudal fin due
to potential fin erosion.
The 29 669 lumpfish observations were obtained after filtering the
original data based on various criteria. The original data set contained
32 641 observations of lumpfish. For example, before November 3rd,
2022, lumpfish was included as a default option for cleaner fish species
in the digital form. Hence, there were occurrences of wrongly registered
lumpfish in the data. We attempted to filter these out based on a clas-
sification model fitted to the observations. The details on the filtering
are provided in the supplementary material. As shown in Figs S2 and S3,
the species were easily separable from length-weight relationships, and
the classification seems reasonable.
2.2. Data set for total and standard length
The data were collected from a Norwegian salmon farm in Tr ø ndelag,
Norway on August 16th, 2023, as part of the lumpfish routine surveil-
lance. The data contain measurements of weight measured in g, total
and standard length measured in cm, and a scoring of caudal fin erosion
for 93 randomly sampled lumpfish individuals. These data only contain
normal lumpfish and hence not deformed lumpfish. No infectious dis-
eases had been diagnosed on the lumpfish at the salmon farm. The
lumpfish were sampled early during the day, before feeding. The caudal
fin score ranged from 0 to 3, where 0 was defined as the caudal fin being
intact, 1 was defined as mildly eroded, 2 was defined as clearly eroded,
and 3 was defined as completely eroded. In the data, there were 39
lumpfish with caudal fin score 0, 45 lumpfish with caudal fin score 1, 9
lumpfish with caudal fin score 2, and 0 lumpfish with caudal fin score 3.
S. Engebretsen et al.
Aquaculture Reports 35 (2024) 101996
3
All the lumpfish with caudal fin erosion were healed, and the damage
had likely occurred during vaccination/handling of the lumpfish during
stocking (hence not after deployment in the sea cages). A picture illus-
trating total and standard length is provided in Fig. 1 .
3. Method
3.1. Condition factor
A general expression for the condition factor, K , is given as.
K = a ⋅ W ∕ L b ,
where L is the length of the fish, W is the weight of the fish, the
parameter b controls how weight increases with length, and the
parameter a is a proportionality constant. For Fulton ’ s K, b = 3, implying
that the shape of the fish is constant with length. For Gutierrez-Rabadan
et al. (2021) , the estimated b post-deployment was 2.559. In this study,
we will set a so that the mean condition factor for our observations is 1
for all the different condition factors.
We obtain an estimate for b based on the 29 669 lumpfish observa-
tions by a linear regression model as
log W = β 0 + b log L + ϵ ,
where ϵ is a random noise term, assumed to be normally distributed with
mean zero. This can alternatively be written as
W ∕ L b = exp ( β 0 + ϵ ) ,
which shows that the proposed condition factor (left hand side of the
equation) by definition varies around a constant level which is inde-
pendent of length. The model is fitted by ordinary least squares. Note
that since the model is fitted on logarithmic scale, the estimated b will be
such that the correlation between the logarithm of the condition factor
and the logarithm of length is 0.
3.2. Condition factor for total length
As total lengths for lumpfish are commonly measured and reported,
we also provide a condition factor which can be used when length is
measured as total length, K
T
. This is done by assuming that total length is
proportional to standard length, and estimating the proportionality
factor by taking the average over the 93 individual ratios between total
and standard length for the lumpfish. Hence, we assume that the fin is a
constant proportion of the total length.
3.2.1. Caudal fin erosion
We also compute the proportionality constant between total and
standard length stratified by the different caudal fin scores. Moreover,
we compare the relationship of caudal fin erosion and condition
measured using total length and standard length. The relationship is
assessed both visually and by fitting a linear regression model to study
whether the condition factor (explanatory variable) can explain the
caudal fin score (response variable). In the regression model, we assume
the same difference between all the levels of caudal fin erosion. We
hence fit the following regression models
CS = β 0 + β 1 ⋅ K + ϵ ,
and
CS = β t
0 + β t
1 ⋅ K T + ϵ t ,
where CS is the caudal fin score, ϵ and ϵ
t
are assumed to be vectors of
independent, mean-zero normally distributed variables, and we esti-
mate the regression parameters β 0 , β t
0 , β 1 , and β t
1 by ordinary least
squares. We are interested in the estimated β
1
and β t
1 .
4. Results
4.1. Estimated condition factor
We obtained the following expression for the condition factor
Proposed K = 5 . 85⋅ W ∕ L 2 . 5016 .
The R
2
obtained from the fitted regression model between weight and
length was 0.79. A histogram of the resulting distribution of condition
factors in the data is provided in Fig. 2 . A selection of observed quantiles
is provided in Table 1 . Note again that the proportionality constant had
been chosen so that the mean condition factor was 1.
In the supplementary material, we also include a sensitivity analysis,
where we fitted the model on the subset of observations with lengths
corresponding to expected weight between 20 g and 250 g according to
the presently proposed condition factor (6.7 cm and 18.4 cm). The
resulting coefficients were similar.
Fig. 1. Standard and total length. The definition of standard and total length of lumpfish. The distance between the two blue lines marks the standard length, while
the total length is the distance between the leftmost blue and the green line.
S. Engebretsen et al.
Aquaculture Reports 35 (2024) 101996
4
4.2. Condition factor versus length
In Fig. 3 a and c, we have plotted our proposed condition factor and
Fulton ’ s K versus logarithm of length, respectively. For Fulton ’ s K, there
was a negative correlation between length and condition factor, so that
larger fish tended to have lower condition factors (correlation − 0.34).
This was not the case for our proposed condition factor (correlation
0.04). This is also clear from Fig. 3 b and d, where we plotted weight
versus length, with the observations coloured by condition factor for our
proposed condition factor and Fulton ’ s K, respectively. For Fulton ’ s K,
all the observations with comparatively high condition factors were for
low lengths. This was not the case for our condition factor, where we had
observations with high condition factor for a wider length range.
4.3. Condition factor for total length
We found a proportionality factor between total length and standard
length of 1.2 (standard deviation 0.03), such that
L T = 1 . 2⋅ L S ,
where L
T
and L
S
are the total and standard length, respectively. The
observed total lengths versus the estimated total lengths, that is, the
standard length times this estimated proportionality factor, are provided
in Fig. 4 . Visually, the assumption of proportionality between the two
length measurements seemed reasonable.
Hence, the corresponding expression for the condition factor for total
length was found to be.
Proposed K T = 9 . 14⋅ W ∕ L 2 . 5016
T ,
where L
T
is the total length measured in cm and 9.14 ≈ 5.85 ⋅ 1. 2
2.5016
(where the approximation is due to rounding of the reported numbers).
4.4. Condition versus caudal fin erosion
We compute the proportionality factor between total and standard
length for the different levels of caudal fin erosion, provided in Table 2 .
From the table, it is clear that caudal fin erosion affects total length, and
hence that this measurement is not appropriate for comparing lumpfish
with and without caudal fin erosion. This can also be seen in Fig. 4 ,
which shows that total length for the lumpfish with caudal fin erosion
was overestimated, and total length for lumpfish without any erosion
was on average underestimated.
We plotted the condition factor for standard and total length versus
caudal fin erosion scores for the 93 lumpfish in our data, provided in
Fig. 5 . Visually, we noted that for our proposed condition factor based on
standard length, there was a negative relationship between condition
factor and caudal fin erosion. This relationship was not found when
length was measured as total length.
We fitted the regression models for caudal fin erosion versus condi-
tion factor for both standard and total length. For standard length, the
estimated regression coefficient was
β 1 = − 0 . 669, with a corre-
sponding estimated p-value of 0.06. For total length, the estimated
regression coefficient was
β t
1 = 0 . 0323, with a corresponding estimated
p-value of 0.9. Hence, we found a tendency for lumpfish with caudal fin
erosion to be of comparatively lower body condition when condition
was measured by standard length, while we did not find this relationship
for total length.
5. Discussion
Based on almost 30 000 lumpfish sampled from Norwegian salmon
farms during production, we have estimated a condition factor for
lumpfish. We found lumpfish growth to be allometric, where weight was
estimated to grow with length raised to the power of 2.50, i.e.
b = 2 . 50.
This is much lower than b = 3 assumed for Fulton ’ s K. When b < 3, we
have that the fish become thinner with increasing length, while for b >
3, the fish become progressively thicker. A meta-analysis ( Froese, 2006 )
of condition factors for many different species found b to be in the range
of 2.5 – 3.5. The only two species with sufficient data which were found
to have b outside of this range were the red bandfish Cepola macro-
pthalma ( b = 2.0) and the blackfin icefish Chaenocephalus aceratus ( b =
3.7). Another species pointed out as allometric is the king soldier-bream
Argyrops spinifer with b = 2.5, similar to our estimate. Hence, as our
estimate was in the lower range of these estimates, it indicates that
lumpfish deployed in salmonid farms were found to be particularly
allometric.
As lumpfish growth is far from isometric, it is not surprising that
Fulton ’ s K is inappropriate for comparing body condition of lumpfish.
We have demonstrated this by showing that for Fulton ’ s K, lumpfish
with longer lengths tended to have lower condition factors. All the
lumpfish with comparatively high body condition were among those
with lowest observed lengths. As the condition factor is not independent
of length for Fulton ’ s K, one needs to know both the length and condi-
tion factor for the lumpfish in order to compare them. However, if one
needs to consider both length and condition factor, then it is not clear
what additional information the condition factor provides, that one
cannot gain from simultaneously considering length and weight.
In the present study, we have aimed for an alternative condition
factor which is tailored to lumpfish. As this condition factor is used as an
easy-to-measure and rough indicator of lumpfish welfare, it needs to be
an overall condition factor. Hence, it may be that it does not fit all
Fig. 2. Proposed condition factor. Distribution of condition factor in our data.
Table 1
Observed quantiles for lumpfish condition factor.
Quantile 2.5% 5% 10% 25% 50% 75% 90% 95% 97.5%
Condition factor 0.508 0.562 0.63 0.768 0.973 1.191 1.395 1.547 1.68
S. Engebretsen et al.
Aquaculture Reports 35 (2024) 101996
5
lengths equally well, as discussed in De Robertis and Williams (2008) . If
the aim had instead been to find the best fitting model for weight based
on length not restricted to this particular shape, then it would be
possible to obtain a better fit.
As this is a study of lumpfish deployed as cleaner fish during
production in salmon farms, it is unlikely that the estimated condition
factor is also appropriate for wild lumpfish in general. For example, the
length-weight relationship likely depends on factors that are different in
salmonid farms and outside farms, like available food and removal of
fish based on visual inspection of weight and/or length.
Fig. 3. Condition factor, length and weight. a) Condition factor versus logarithm of length for our proposed condition factor. b) Weight versus length coloured by our
proposed condition factor. The orange line represents the mean weight as a function of length implied by our proposed condition factor. c) Condition factor versus
logarithm of length for Fulton ’ s K. d) Weight versus length coloured by Fulton ’ s K. The orange line represents the mean weight as a function of length implied by
Fulton ’ s K.
Fig. 4. Total versus standard length. Observed total length versus the estimated proportionality factor times the observed standard lengths.
S. Engebretsen et al.
Aquaculture Reports 35 (2024) 101996
6
5.1. Comparison with Guiterrez-Rabadan ’ s K
Reassuringly, our estimated b in the expression for the condition
factor was very similar to that reported in Gutierrez-Rabadan et al.
(2021) . While we found an estimate of
b = 2 . 502, the estimate in
Gutierrez-Rabadan et al. (2021) was 2.559. Hence, we confirm the re-
sults of Gutierrez-Rabadan et al. (2021) with an independent, larger data
set. Note that in Gutierrez-Rabadan et al. (2021) , the total length was
measured for the lumpfish. As the estimated b ’ s were so similar, these
two condition factors are roughly proportional to each other. Hence, for
comparing condition over time or condition between farms, the two
condition factors would roughly be equally appropriate and in most
settings yield the same qualitative conclusions. However, the interpre-
tation of the value of the body condition is different for the two ex-
pressions. The condition factor and operational welfare indicator
( Boissonnot et al., 2023 , 2022a ) based on Gutierrez-Rabadan et al.
(2021) is used to assess the degree of emaciation. On the other hand, the
presently proposed condition factor is defined so that the average con-
dition among all the 29 669 lumpfish is set to 1. Hence, it can be used to
assess how a specific lumpfish compares to an average lumpfish
deployed in a Norwegian fish farm. Note that we do not define cut-offs
for good condition factor. In order to do that, it is necessary to further
analyse the expression and compare it to multiple nutritional and wel-
fare parameters. Our condition factor is defined such that the average is
1, independent of length. This does not necessarily mean that a normal,
healthy fish will have condition 1 over time. It could be that fish become
gradually more emaciated over time in sea cages. It could also be that
they become gradually fatter, for example as they start feeding on
salmon pellets.
5.2. Caudal fin erosion, total and standard length
Standard length was the most appropriate length measurement for
measuring body condition and comparing lumpfish. This was because
for total length, lumpfish with eroded fins will have comparatively
higher condition than lumpfish with their fins intact. We found that the
relationship between total and standard length was dependent on caudal
fin erosion, implying that total length is not comparable between
lumpfish with different levels of caudal fin erosion. We found a tendency
for higher degrees of caudal fin erosion among lumpfish with lower
condition based on standard length (p-value 0.06). For total length, the
estimate was in the opposite direction (p-value 0.9). This indicates that
when condition factor is computed based on total length, it cannot be
used to compare lumpfish with different levels of caudal fin erosion.
Damage on the caudal fin of lumpfish is a common problem, and was for
example found for 42.5% of the lumpfish studied in ( Boissonnot et al.,
2023 ), though most of these only had slight damage. In the data set used
in the present study, 58% were found to have damage on the caudal fin.
5.3. Limitations
As there were clearly human errors in the registered observations, we
had to filter out observations based on their weight and length. Hence, it
could be that we have filtered out real observations which were extreme
and rare. It could also be that we have wrongly included wrasse obser-
vations with atypical weight-length relationships for wrasse. However,
given the amount of data, this should not qualitatively affect our results.
5.4. Future work
With an appropriate measure of body condition, we can compare
plumpness of lumpfish between different fish farms or over time, as an
indicator of welfare. However, there may be other factors which also
affect lumpfish body condition. For other species, the condition factor
has been found to vary with for example seasonal fluctuations in
metabolic balance, gonad development, age, sex, and reproduction
timing ( De Robertis and Williams, 2008; Blackwell et al., 2000; Bolger
and Connolly, 1989; Froese, 2006 ). Length-weight relationships may
Table 2
Proportionality factor between total length and standard length for the different
levels of caudal fin erosion.
Caudal fin score 0 1 2
Number of lumpfish 39 45 9
Proportionality factor 1.21 1.19 1.15
Fig. 5. Condition factor, total and standard length and caudal fin erosion. Condition factor versus caudal fin erosion for a) The proposed condition factor based on
standard length. b) The proposed condition factor based on total length.
S. Engebretsen et al.
Aquaculture Reports 35 (2024) 101996
7
also vary with different strains of lumpfish. These factors may affect the
comparability of lumpfish body condition over time and between fish
farms. Hence, it is a topic for future work to identify potential
non-operational factors which may affect the length-weight relationship
of lumpfish.
As body condition only requires measurements of length and weight
of the fish, it is a low-cost measure which is minimally invasive for the
lumpfish. We therefore recommend using lumpfish condition factor as
part of the total welfare surveillance in fish farms during production.
Though the condition factor can be used to measure body condition (i.e.
fatness of the fish), it is not clear to what degree it can be used to assess
nutritional status in general. We did, however, find an indication of a
negative association between body condition and caudal fin erosion, i.e.
that eroded caudal fins were less prevalent among lumpfish with
comparatively higher body condition. Boissonnot et al. (2022b) studied
the association between body condition and the amount of fat vacuoli-
sation in livers. Similarly, future studies should extend this work and
compare body condition of lumpfish with nutritional and welfare mea-
sures, like protein biomarkers or simpler, visual welfare indicators like
cataracts or other types of fin erosion. Such studies can also be used to
assess what condition factors can be considered as good or bad
condition.
Having established a measure for lumpfish body condition, it can be
used to study how body condition depends on different feeding regimes
and other controllable, operational conditions and farming strategies.
Future work should thus study different controllable farming practices
and their effect on body condition, which can in turn be used as concrete
input to best practice guides for fish farmers.
Funding
This work was supported by the Norwegian Seafood Research Fund
[Rensefiskbetingelser project 901766; STRATEGI project 901693],
Bj ø r ø ya AS[project CycLus, NTF36-37] and the Norwegian Research
Council [SkatteFUNN project 260305].
CRediT authorship contribution statement
Solveig Engebretsen : Formal analysis; Investigation; Methodology;
Conceptualization; Visualization; Writing - original draft; Writing - re-
view & editing. Magne Aldrin : Formal analysis; Investigation; Meth-
odology; Conceptualization; Visualization; Writing - original draft;
Writing - review & editing. Liss Lunde : Data curation; Investigation;
Writing - review & editing. Marthe Austad : Conceptualization; Writing
- review & editing. Trond Rafoss : Data curation; Conceptualization;
Writing - review & editing. Ole Roald Danielsen : Data curation;
Conceptualization; Writing - review & editing. Andreas Lindhom : Data
curation; Conceptualization; Writing - review & editing. Lauris Bois-
sonnot : Conceptualization; Writing - review & editing. Peder A. Jan-
sen : Investigation; Methodology; Conceptualization; Writing - original
draft; Writing - review & editing.
Declaration of Competing Interest
The authors declare the following financial interests/personal re-
lationships which may be considered as potential competing interests:
Andreas Lindhom reports a relationship with Norsk Oppdrettsservice AS
that includes: employment and equity or stocks. Ole Roald Danielsen
reports a relationship with Norsk Oppdrettsservice AS that includes:
employment. Trond Rafoss reports a relationship with Landbasert
Akvakultur Norge AS that includes: equity or stocks.
Data availability
The authors do not have permission to share data.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the
online version at doi:10.1016/j.aqrep.2024.101996 .
References
Barrett, L.T., Overton, K., Stien, L.H., Oppedal, F., Dempster, T., 2020. Effect of cleaner
fish on sea lice in norwegian salmon aquaculture: a national scale data analysis. Int.
J. Parasitol. 50, 787 – 796. https://doi.org/10.1016/j.ijpara.2019.12.005 .
Blackwell, B.G., Brown, M.L., Willis, D.W., 2000. Relative weight (w r ) status and current
use in fisheries assessment and management. Rev. Fish. Sci. 8, 1 – 44. https://doi.org/
10.1080/10641260091129161 .
Boissonnot, L., Austad, M., Karlsen, C., Reynolds, P., Stensby-Skjærvik, S., 2022a.Welfare
assessment of lumpfish in sea cages - Manual. 〈 https://aqua-kompetanse.no/lu
mpfish/ 〉 .version 2.
Boissonnot, L., Engebretsen, S., Aldrin, M., 2022b. Kondisjonsfaktor hos rognkjeks. Nor.
Fiskeoppdrett 4, 62 – 65 .
Boissonnot, L., Karlsen, C., Reynolds, P., Austad, M., Stensby-Skjærvik, S., Remen, M.,
Imsland, A.K., 2023. Welfare and survival of lumpfish (Cyclopterus lumpus) in
norwegian commercial atlantic salmon (Salmo salar) production. Aquaculture 572,
739496. https://doi.org/10.1016/j.aquaculture.2023.739496 .
Bolger, T., Connolly, P., 1989. The selection of suitable indices for the measurement and
analysis of fish condition. J. Fish. Biol. 34, 171 – 182. https://doi.org/10.1111/
j.1095-8649.1989.tb03300.x .
Brooker, A.J., Papadopoulou, A., Gutierrez, C., Rey, S., Davie, A., Migaud, H., 2018.
Sustainable production and use of cleaner fish for the biological control of sea lice:
recent advances and current challenges. Vet. Rec. 183, 383. https://doi.org/
10.1136/vr.104966 .
Daborn, G.R., Gregory, R.S., 1983. Occurrence, distribution, and feeding habits of
juvenile lumpfish, Cyclopterus lumpus L. in the Bay of Fundy. Can. J. Zool. 61,
797 – 801. https://doi.org/10.1139/z83-105 .
De Robertis, A., Williams, K., 2008. Weight-length relationships in fisheries studies: the
standard allometric model should be applied with caution. Trans. Am. Fish. Soc. 137,
707 – 719. https://doi.org/10.1577/T07-124.1 .
Eliasen, K., Patursson, E.J., McAdam, B.J., Pino, E., Morro, B., Betancor, M., Baily, J.,
Rey, S., 2020. Liver colour scoring index, carotenoids and lipid content assessment as
a proxy for lumpfish ( Cyclopterus lumpus L.) health and welfare condition. Sci. Rep.
10, 1 – 12. https://doi.org/10.1038/s41598-020-65535-7 .
Engebretsen, S., Aldrin, M., Qviller, L., Stige, L.C., Rafoss, T., Danielsen, O.R.,
Lindhom, A., Jansen, P.A., 2023. Salmon lice (Lepeophtheirus salmonis) in the
stomach contents of lumpfish (Cyclopterus lumpus) sampled from norwegian fish
farms: relationship between lice grazing and operational conditions. Aquaculture
563, 738967. https://doi.org/10.1016/j.aquaculture.2022.738967 .
Fiskeridirektoratet, 2022.Utsett av rensefisk 1998-2021 - publ. 13.10.2022. 〈 https
://www.fiskeridir.no/Akvakultur/Tall-og-analyse/Akvakulturstatistikk-tidsserier/
Rensefisk/ 〉 .14-Feb-2023.
Flores-García, L., Camargo-Castellanos, J.C., Pascual-Jímenez, C., Almaz ´
an-Rueda, P.,
Monroy-L ´
opez, J.F., Albertos-Alpuche, P.J., Martínez-Y ´
a ˜
nez, R., 2022. Welfare
indicators in tilapia: an epidemiological approach. Front. Vet. Sci. 9. https://doi.org/
10.3389/fvets.2022.882567 .
Froese, R., 2006. Cube law, condition factor and weight – length relationships: history,
meta-analysis and recommendations. J. Appl. Ichthyol. 22, 241 – 253. https://doi.
org/10.1111/j.1439-0426.2006.00805.x .
Geitung, L., Wright, D.W., Oppedal, F., Stien, L.H., Vågseth, T., Madaro, A., 2020.
Cleaner fish growth, welfare and survival in atlantic salmon sea cages during an
autumn-winter production. Aquaculture 528, 735623. https://doi.org/10.1016/j.
aquaculture.2020.735623 .
Gutierrez-Rabadan, C., Spreadbury, C., Consuegra, S., de Leaniz, C.G., 2021.
Development, validation and testing of an Operational Welfare Score Index for
farmed lumpfish Cyclopterus lumpus L. Aquaculture 531, 735777. https://doi.org/
10.1016/j.aquaculture.2020.735777 .
Imsland, A.K., Reynolds, P., Eliassen, G., Hangstad, T.A., Foss, A., Vikingstad, E.,
Elvegård, T.A., 2014. The use of lumpfish ( Cyclopterus lumpus L.) to control sea lice
( Lepeophtheirus salmonis Kr ø yer) infestations in intensively farmed Atlantic salmon
( Salmo salar L.). Aquaculture 424, 18 – 23. https://doi.org/10.1016/j.
aquaculture.2013.12.033 .
Imsland, A.K., Reynolds, P., Eliassen, G., Hangstad, T.A., J ´
onsd ´
ottir, O.D., Elvegård, T.A.,
Lemmens, S.C., Rydland, R., Nytr ø , A.V., 2016. Investigation of behavioural
interactions between lumpfish ( Cyclopterus lumpus ) and goldsinny wrasse
( Ctenolabrus rupestris ) under controlled conditions. Aquac. Int. 24, 1509 – 1521.
https://doi.org/10.1007/s10499-016-0008-y .
Imsland, A.K., Reynolds, P., Hangstad, T.A., J ´
onsd ´
ottir, ´
O.D., Noble, T., Wilson, M.,
Mackie, J.A., Elvegård, T.A., Urskog, T.C., Mikalsen, B., 2018a. Feeding behaviour
and growth of lumpfish ( Cyclopterus lumpus L.) fed with feed blocks. Aquac. Res. 49,
2006 – 2012. https://doi.org/10.1111/are.13657 .
Imsland, A.K., Reynolds, P., Jonassen, T.M., Hangstad, T.A., Elvegård, T.A., Urskog, T.C.,
Hanssen, A., Mikalsen, B., 2019. Effects of different feeding frequencies on growth,
cataract development and histopathology of lumpfish ( Cyclopterus lumpus L.).
Aquaculture 501, 161 – 168. https://doi.org/10.1016/j.aquaculture.2018.11.026 .
Imsland, A.K., Reynolds, P., Jonassen, T.M., Hangstad, T.A., Elvegård, T.A., Urskog, T.C.,
Mikalsen, B., 2018b. Effects of three commercial diets on growth, cataract
development and histopathology of lumpfish ( Cyclopterus lumpus L.). Aquac. Res. 49,
3131 – 3141. https://doi.org/10.1111/are.13776 .
S. Engebretsen et al.
Aquaculture Reports 35 (2024) 101996
8
Imsland, A.K., Reynolds, P., Lorentzen, M., Eilertsen, R.A., Micallef, G., Tvenning, R.,
2020. Improving survival and health of lumpfish ( Cyclopterus lumpus L.) by the use of
feed blocks and operational welfare indicators (OWIs) in commercial Atlantic
salmon cages. Aquaculture 527, 735476. https://doi.org/10.1016/j.
aquaculture.2020.735476 .
Imsland, A.K.D., Berg, M.S., Haugland, G.T., Eliasen, K., 2022. Comparing body density
of lumpfish ( Cyclopterus lumpus ) to different operational welfare indicators. Fishes 7,
284. https://doi.org/10.3390/fishes7050284 .
Klakegg, Ø ., Myhren, S., Juell, R.A., Aase, M., Salonius, K., S ø rum, H., 2020. Improved
health and better survival of farmed lumpfish ( Cyclopterus lumpus ) after a probiotic
bath with two probiotic strains of Aliivibrio. Aquaculture 518, 734810. https://doi.
org/10.1016/j.aquaculture.2019.734810 .
Le Cren, E.D., 1951. The length-weight relationship and seasonal cycle in gonad weight
and condition in the perch ( Perca fluviatilis ). J. Anim. Ecol. 201 – 219. https://doi.
org/10.2307/1540 .
Garcia de Leaniz, C., Gutierrez Rabadan, C., Barrento, S.I., Stringwell, R., Howes, P.N.,
Whittaker, B.A., Minett, J.F., Smith, R.G., Pooley, C.L., Overland, B.J., et al., 2022.
Addressing the welfare needs of farmed lumpfish: Knowledge gaps, challenges and
solutions. Rev. Aquac. 14, 139 – 155. https://doi.org/10.1111/raq.12589 .
Nash, R.D., Valencia, A.H., Geffen, A.J., 2006. The origin of Fulton ’ s condition
factor — setting the record straight. Fisheries 31, 236 – 238 .
Powell, A., Treasurer, J.W., Pooley, C.L., Keay, A.J., Lloyd, R., Imsland, A.K., Garcia de
Leaniz, C., 2018. Use of lumpfish for sea-lice control in salmon farming: challenges
and opportunities. Rev. Aquac. 10, 683 – 702. https://doi.org/10.1111/raq.12194 .
Ricker, W.E., 1975. Computation and interpretation of biological statistics of fish
populations. Bull. Fish. Res. Can. 191, 1 – 382 .
Scholz, F., Ruane, N., Marcos-Lopez, M., Mitchell, S., Bolton-Warberg, M., O ’ Connor, I.,
Rodger, H., 2018. Systemic mycoses in lumpfish ( Cyclopterus lumpus L.) in Ireland:
aetiology and clinical presentation. Bull. Eur. Assoc. Fish. Pathol. 38, 203 .
S. Engebretsen et al.
