Content uploaded by Luca Lanteri
Author content
All content in this area was uploaded by Luca Lanteri on Feb 19, 2016
Content may be subject to copyright.
Report of the Horse Mackerel,
Mediterranean Horse Mackerel and
Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T.
picturatus) Otolith Exchange 2015
Mahé K., Jurado A., Garcia Guerreiro A., Massaro A., Dueñas C., Lopez E., Mullins E.,
Lanteri L., Ferreira M.J., Elleboode R., Mannini A., Antolinez A., Delfs G., Casciaro L.,
O'Cuaig M., Torres P., Dijkman A., Bellamy E., Eriksen K., Carbonara P., 2015. Report
of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel
(Trachurus trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015. 30pp.
Table of contents
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
1. Introduction .................................................................................. 4
2. Participants .................................................................................. 4
3. Sampling collection ..................................................................... 5
4. Reading procedure ...................................................................... 6
5. Results .......................................................................................... 6
5.1. Precision ............................................................................................................ 7
5.1.1. Trachurus mediterraneus ...................................................................... 7
5.1.2. Trachurus picturatus ............................................................................. 7
5.1.3. Trachurus trachurus .............................................................................. 8
5.2. Relative bias (Accuracy) .................................................................................... 8
5.2.1. Trachurus mediterraneus ...................................................................... 8
5.2.2. Trachurus picturatus ............................................................................. 9
5.2.3. Trachurus trachurus .............................................................................. 9
6. Images of reference ..................................................................... 10
6.1. Trachurus trachurus ........................................................................................... 10
6.2. Trachurus picturatus .......................................................................................... 13
6.3. Trachurus mediterraneus ................................................................................... 13
7. Abstract ........................................................................................ 14
8. References .................................................................................... 15
Appendix 1 : List of participants .................................................... 16
Appendix 2: Details results of Trachurus trachurus ..................... 18
Appendix 3: Details results of Trachurus picturatus .................... 22
Appendix 4: Details results of Trachurus mediterraneus ............ 26
4
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
1. Introduction
The Planning Group on Commercial Catch, Discards and Biological Sampling
(PGCCDBS) meeting in February 2014 recommended a large exchange in 2014 and a
workshop for 2015 (ICES, 2014). The planning group indicated that Kelig Mahe
(IFREMER, France) and Pierluigi Carbonara (COISPA Tecnologia & Ricerca, Italy)
will be responsible to organising a Trachurus species (Trachurus trachurus, T.
mediterreaneus and T. picturatus) otolith exchange and workshop. There were 3 age
reading workshops (1999, 2006 and 2012) which were preceded by one exchange.
2. Participants
19 readers from 8 countries participated at this exchange (Tab. 1).
Table 1: List of the readers.
Reader Name Country Institution
1 Alba Jurado Spain Instituto Español de Oceanografia
2 Alexandra Garcia
Guerreiro Portugal IMAR/University of Azores
3 Andrea Massaro Italy CIBM
4 Clara Dueñas Spain Instituto Español de Oceanografia
5 Eduardo Lopez Spain Instituto Español de Oceanografia
6 Eugene Mullins Ireland Marine Institute
7 Luca Lanteri Italy University of Genoa
8 Maria João Ferreira Portugal IPMA
9 Romain Elleboode France Institut Français de Recherche pour
l'Exploitation de la Mer
10 Alessandro Mannini Italy Italian Society for Marine Biology
11 Ana Antolinez Spain Instituto Español de Oceanografia
12 Gertrud Delfs Germany Thünen Institute of Sea Fisheries
13 Loredana Casciaro Italy COISPA
14 Macdara O'Cuaig Ireland Marine Institute
15 Pedro Torres Spain Instituto Español de Oceanografia
16 Andre Dijkman Netherland
s Wageningen IMARES
17 Elise Bellamy France Institut Français de Recherche pour
l'Exploitation de la Mer
18 Pierluigi Carbonara Italy COISPA
19 Kirsti Eriksen Norway Institute of Marine Research
Appendix 1 presents the complete listing of the participants in the Trachurus species
otolith exchange.
Sampling collection 5
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
3. Sampling collection
A total of 550 fish was sampled from 2003 to 2014 (2003, 2004, 2005, 2006, 2007,
2009, 2011 and 2014) from both Atlantic and Mediterranean area (Tab. 2):
95 Trachurus mediterraneus sampled in the three geographical areas by
COISPA (Italy) and IEO (Spain)
134 Trachurus picturatus sampled in the three geographical areas by IEO
(Spain), COISPA (Italy) and DOP (Portugal)
321 Trachurus trachurus sampled in the three geographical areas by IFREMER
(France), DISTAV (Italy), TI-SF (Germany) and IEO (Spain)
Table 2: Samples distribution by Trachurus species (number corresponding to map).
Species
Atlantic Ocean Mediterranean Sea
Total
Eastern
Channel
VIId
(1)
Celtic
Sea
VIIh
(2)
Bay of
Biscay
VIIIc
(3)
Azores
XII
(4)
Portugu
ese
waters
IXa
(5)
Tenerife
(6)
Alboran
Sea
(7)
South
Adriatic
Sea
(8)
Ligurian
Sea
(9)
T.
mediterraneus 35 10 50 95
T. picturatus 71 50 13 134
T. trachurus 50 154 20 97 321
1
2
3
4
6
5
7
9
8
Figure 1: Map of Trachurus species sampling areas (Source: Google Map 2015).
Reading procedure 6
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
The samplings came from both the harbor and the survey. The otolith sections and
whole otoliths were used during the 2014/2015 exchange but only Trachurus trachurus
was sampled with both techniques (Table 3).
Table 3: Samples distribution by Trachurus species and by preparation method of otolith.
Species
Otolith
section
Whole
otolith
Trachurus mediterraneus
95
Trachurus picturatus
134
Trachurus trachurus
201
120
Total
154
349
4. Reading procedure
One image of each otolith was uploaded to WebGR
(http://webgr.azti.es/ce/search/myce). All participants received all informations to
participate to this exercise in the WebGR tool. The WebGR tool was used to this
exchange. The use of WebGR tool for the exchange has somes advantages: (i) it can
facilitate and accelerate the whole exchange process, (ii) annotated images are obtained
for every otolith which enables to compare age readings directly and to identify possible
sources of bias (iii) it is very easy for the chairman to compile the results. However, the
use of WebGR tool for the exchange present some limits: (i) the WebGR tool is not very
intuitive tool (ii) the WebGR could be jammed (iii) it is not always possible to upload a
large batch of images (problem with the format of the csv file with Windows 7).
The age was assigned taking into account the number of the transparent rings. Moreover
the date of capture, the sex and total length were visible by the readers.
All data were extracted of the WebGR and after agreements between age-readers were
calculated using the Guus Eltink spreadsheet (Eltink, 2000).
5. Results
The spreadsheet (Eltink, 2000) was completed according to the instructions contained in
Guidelines and Tools for Age Reading Comparisons by Eltink et al. (2000). Modal ages
were calculated for each otolith red, with percentage agreement, mean age and precision
coefficient of variation as a definition (for each otolith):
percentage agreement = 100x(no. of readers agreeing with modal age/total no. of
readers).
precision c. v. = 100x(standard deviation of age readings/mean of age readings).
Results 7
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
5.1. Precision
The precision
1
analyse with Coefficient of Variation (CV) and percent of agreement to
modal age for otoliths sets according to the Trachurus species was presented in the
Table 4. Among three Trachurus species, all data showed the very low precision with
the percent agreement between 47 and 56% and the CV from 29 to 69.
Table 4: Reading’s precision by Trachurus species.
Trachurus
species
Otoliths
number
Readers
number
Age
range
Percentage of
Agreement
CV
T. mediterraneus
95 15 0/12 47.1% 43.8%
T. picturatus
133 13 0/15 48.9% 69.0%
T. trachurus
309 17 0/15 55.8% 28.7%
5.1.1. Trachurus mediterraneus
The results of otoliths readings of Trachurus mediterraneus showed a better precision in
the Mediterranean area than those obtained in the sampling areas from the Atlantic
Ocean (Tab. 5).
Table 5: Reading’s precision of Trachurus mediterraneus
by sampling areas
.
Sampling
areas
Otoliths
number
Readers
number Age range
Percentage of
Agreement CV
VIIIc
35 15 0/10 39.3% 40.2%
IXa
10 15 0/6 41.2% 41.7%
South Adriatic
Sea
50 15 0/12 53.6% 46.7%
5.1.2. Trachurus picturatus
The results of otoliths readings of Trachurus picturatus showed a lower precision in the
Azores area than those obtained in the other sampling areas due to especially the old
specimen in this area (Tab. 6).
Table 6: Reading’s precision by Trachurus
picturatus by sampling areas
.
Sampling
areas
Otoliths
number
Readers
number Age range
Percentage of
Agreement CV
Azores
70 13 0/15 35.3% 36.0%
Tenerife
50 12 0/5 60.1% 89.3%
South Adriatic
Sea
13 12 0/1 79.3% 168.8%
1
Precision is defined as the variability in the age readings. The precision's errors in age readings are
better described by the coefficient of variation (CV) by age group. This measure of precision is
independent of the closeness to the true age (ICES, 2007).
Results 8
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
5.1.3. Trachurus trachurus
The results of otoliths readings of Trachurus trachurus showed a better precision in the
Atlantic Ocean (VIId and VIIh) than those obtained in the sampling areas from the
Mediterranean Sea (Alboran and Ligurian sea) (Tab. 7). Moreover, in the Ligurian Sea,
the analysis between otoliths section and whole otoliths showed the same level of
precision (Tab. 7).
Table 7: Reading’s precision by Trachurus
trachurus by sampling areas
.
Sampling
areas
Otoliths
number
Otolith
preparation
Readers
number
Age
range
Percentage of
Agreement
CV
VIId
50 Section 16 2/15 55.7% 16.8%
VIIh
154 Section 16 1/14 63.8% 25.9%
Alboran Sea
20 Whole 17 1/3 50.1% 69.7%
Ligurian Sea
45 Section 13 1/14 44.6% 32.6%
Ligurian Sea
50 Whole 14 1/10 44.0% 28.9%
5.2. Relative bias (Accuracy)
2
The minimal requirement for age reading's consistency is the absence of bias among
readers and through time. The hypothesis of an absence of bias between two readers or
between a reader and the modal age estimated can be tested non-parametrically with a
one-sample Wilcoxon signed rank test.
5.2.1. Trachurus mediterraneus
It should be noted that there are certainly of bias between some readers and modal age
for Trachurus mediterraneus otoliths. There were no observed bias between five readers
(5/15 readers, 33%) and the modal age (Tab. 8).
2
In absence of calcified structures of known age, the age readings can be compared to modal age, which
is defined as the age determined for an individual structure whose most of the readers have a preference.
Relative bias can be defined as a systematic over- or underestimation of age compared to the modal age.
The age reading comparisons to modal age provide a low estimate of relative bias compared to absolute
bias, when most readers have a similar serious bias in age reading (ICES, 2007).
Results 9
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
Table 8: Inter-reader bias test and reader against modal age bias test of Trachurus mediterraneus otoliths
(-: no sign of bias (p>0.05); *: possibility of bias (0.01<p<0.05); **: certainty of bias (p<0.01)).
5.2.2. Trachurus picturatus
It should be noted that there are certainly of bias between some readers and modal age
for Trachurus picturatus otoliths. There were no observed bias between two readers
(2/13 readers, 15%) and the modal age (Tab. 9).
Table 9: Inter-reader bias test and reader against modal age bias test of Trachurus mediterraneus otoliths
(-: no sign of bias (p>0.05); *: possibility of bias (0.01<p<0.05); **: certainty of bias (p<0.01)).
5.2.3. Trachurus trachurus
It should be noted that there are certainly of bias between some readers and modal age
for Trachurus mediterraneus otoliths. There were no observed bias between five readers
(5/17 readers, 29%) and the modal age (Tab. 10).
Images of reference 10
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
Table 10: Inter-reader bias test and reader against modal age bias test of Trachurus trachurus otoliths (-:
no sign of bias (p>0.05); *: possibility of bias (0.01<p<0.05); **: certainty of bias (p<0.01)).
6. Images of reference
Seven otoliths (7/550, 1.3%) presented 100% agreement.
6.1. Trachurus trachurus
Among all otoliths of Trachurus trachurus (n=309), there were only five otoliths
presented 100% agreement between readers. These five otoliths came from VIIh area
(Celtic Sea) and the observed age was 1 year old.
Figure 2 : Trachurus trachurus otolith image (EB_14_B6_C4_Or_0004.jpg) from the Celtic Sea (VIIh)
with 100% of agreement, 1 year old (from 16 readers). Total length was 10 cm. This specimen caught
April 2014.
Images of reference 11
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
Figure 3 : Trachurus trachurus otolith image (EB_14_B6_C5_OCr_0001.jpg) from the Celtic Sea (VIIh)
with 100% of agreement, 1 year old (from 16 readers). Total length was 10 cm. This specimen caught
April 2014.
Figure 4 : Trachurus trachurus otolith image (EB_14_B6_C5_OCr_0002.jpg) from the Celtic Sea (VIIh)
with 100% of agreement, 1 year old (from 16 readers). Total length was 11 cm. This specimen caught
April 2014.
Images of reference 12
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
Figure 5 : Trachurus trachurus otolith image (EB_14_B6_C5_OCr_0003.jpg) from the Celtic Sea (VIIh)
with 100% of agreement, 1 year old (from 16 readers). Total length was 11 cm. This specimen caught
April 2014.
Figure 6 : Trachurus trachurus otolith image (EB_14_B6_C5_Or_0001.jpg) from the Celtic Sea (VIIh)
with 100% of agreement, 1 year old (from 16 readers). Total length was 10 cm. This specimen caught
April 2014.
Images of reference 13
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
6.2. Trachurus picturatus
Among all otoliths of Trachurus picturatus (n=132), there were only two otoliths
presented 100% agreement between readers. These two otoliths came from South
Adriatic Sea and the observed age was 0 year old.
Figure 7 : Trachurus picturatus otolith image (COISPA – 4967.jpg) from South Adriatic sea with 100%
of agreement, 1 year old (from 13 readers). Total length was 10 cm. This specimen caught July 2013.
Figure 8 : Trachurus picturatus otolith image (COISPA – 4973.jpg) from South Adriatic sea with 100%
of agreement, 0 year old (from 13 readers). Total length was 11 cm. This specimen caught July 2013.
6.3. Trachurus mediterraneus
Among all otoliths of Trachurus mediterraneus (n=95), there were no otoliths presented
100% agreement between readers.
Abstract 14
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
7. Abstract
The ICES Planning Group on Commercial Catch, Discards and Biological
Sampling (PGCCDBS) identified the need of the Horse Mackerel, Mediterranean
Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T. mediterreaneus
and T. picturatus) otolith exchange to take place in 2015. It was the forth exchange.
The IFREMER institute and the COISPA Tecnologia & Ricerca, Italy, coordinated
this exchange. A total of 550 fish was sampled from the Atlantic Ocean (Eastern
Channel, Celtic Sea, Bay of Biscay, Azores, Portuguese waters and Tenerife) and
the Mediterranean Sea (Alboran Sea, South Adriatic Sea and Ligurian Sea). 19
readers from 8 countries (France, Germany, Spain, Ireland, Italy, Portugal,
Netherlands and Norway) participated to this exchange. Among three Trachurus
species, all data showed the very low precision with the percent agreement between
47 and 56% and the CV from 29 to 69%. The precision analysis showed the same
level of precision between otolith sections and whole otoliths from the Ligurian
Sea.
References 15
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
8. References
Eltink, A. T. G. W., Newton, A. W., Morgado, C., Santamaria, M. T. G., Modin, J.,
2000. Guidelines and Tools for Age Reading. (PDF document version 1.0 October
2000) Internet : http://www.efan.no
Eltink, A. T. G. W., 2000. Age reading comparisons. (MS Excel workbook version 1.0
October 2000) Internet : http://www.efan.no
ICES. 2005. Report of the Planning Group on Commercial Catch, Discards and
Biological Sampling (PGCCDBS), 1-4 March 2005, Oostende, Belgium. ICES CM
2005/ACFM:15. 149 pp.
ICES. 2006. Report of the Planning Group on Commercial Catch, Discards and
Biological Sampling (PGCCDBS), 28 February–3 March 2006, Rostock, Germany.
ICES CM 2006/ACFM:18. 62 pp.
ICES. 2007. Report of the Planning Group on Commercial Catch, Discards and
Biological Sampling (PGCCDBS), 5–9 March 2007, Valetta, Malta. ACFM:09. 115p.
ICES. 2014. Report of the Planning Group on Commercial Catches, Discards and Bio-
logical Sampling (PGCCDBS), 17–21 February 2014, Horta (Azores), Portugal. ICES
CM 2014 / ACOM: 34. 103 pp.
References 16
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
Appendix 1 : List of participants
Country Participants in
exchange Email expertise level
T.
mediterraneus
expertise
level T.
pitcuratus
expertise
level T.
trachurus
Italy Loredana
Casciaro casciaro@coispa.it L none none
Pierluigi
Carbonara carbonara@coispa.it H none H
Luca Lanteri luca.lanteri@libero.it M none M
Alessandro
Mannini A.Mannini@unige.it M none M
Fulvio
Garibaldi garibaldi.f@libero.it M none M
Andrea Massaro andreamassaro@live.it M L M
France Kélig Mahé kelig.mahe@ifremer.fr none none none
Elise Bellamy Elise.Bellamy@ifremer.fr H H H
Romain
Elleboode romain.elleboode@ifremer.fr H H H
Germany Gertrud Delfs gertrud.delfs@ti.bund.de none none medium
Christoph
Stransky christoph.stransky@ti.bund.de none none none
Jens Ulleweit jens.ulleweit@ti.bund.de none none none
Ireland Eugene Mullins eugene.mullins@marine.ie none none H
Macdara
O'Cuaig macdara.ocuaig@marine.ie none none Medium-
still in
training
Netherlands André Dijkman andre.dijkman@wur.nl none none M
Bolle, Loes loes.bolle@wur.nl none none none
Norway Kirsti Børve
Eriksen kirstibe@imr.no none none
Jane Godiksen jane.godiksen@imr.no none none L
Spain Begoña
Villamor begona.villamor@st.ieo.es none none none
Clara Dueñas clara.duenas@st.ieo.es L (1 year-200
otoliths) none
H (10
years-
25000
otoliths)
Ana Antolinez ana.antolinez@st.ieo.es L (1 year-100
otoliths) none L (1 year-
500
otoliths)
Eduardo Lopez eduardo.lopez@vi.ieo.es M (2 years-
1000 otoliths) none M (2 years-
800
otoliths)
Mª Teresa
García
Santamaría teresa.garcia@ca.ieo.es L M L
Alba Jurado
Ruzafa alba.jurado@ca.ieo.es L M L
Pedro Torres pedro.torres@ma.ieo.es none none none
Jesus costa jesus.acosta@ma.ieo.es L none H
Portugal Alexandra agarcia@uac.pt none M none
References 17
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus
trachurus, T. mediterreaneus and T. pictatus) Otolith Exchange 2015
Garcia
Poland Tomasz Nermer tnermer@mir.gdynia.pl none none H
Denmark Aage Thaarup att@aqua.dtu.dk none none H
Portugal Maria João
Ferreira mjferreira@ipma.pt L (1 year-200
otoliths) L (1 year-200
otoliths)
H (13
years-
26000
otoliths)
18
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
Appendix 2: Details results of Trachurus trachurus
The number of age readings, the coefficient of variation (CV), the percentage of agreement and
the RELATIVE bias are presented by MODAL age for each age reader and for all readers
combined. A weighted mean CV and a weighted mean percent agreement are given by reader and
all readers combined. The CV's by MODAL age for each individual age reader and all readers
combined indicate the precision in age reading by MODAL age. The weighted mean CV's over all
MODAL age groups combined indicate the precision in age reading by reader and for all age
readers combined.
References 19
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
In the age bias plots below the mean age recorded +/- 2stdev of each age reader and all readers
combined are plotted against the MODAL age. The estimated mean age corresponds to MODAL
age, if the estimated mean age is on the 1:1 equilibrium line (solid line). RELATIVE bias is the
age difference between estimated mean age and MODAL age.
References 20
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
21
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
The coefficient of variation (CV%), percentage of agreement and the standard deviation (STDEV)
are plotted against MODAL age. CV is much less age dependent than the standard deviation
(STDEV) and the percentage of agreement. CV is therefore a better index for the precision in age
reading. Problems in age reading are indicated by relatively high CV's at age.
The distribution of the age reading errors in percentage by MODAL age as observed from the
whole group of age readers in an age reading comparison to MODAL age. The achieved precision
in age reading by MODAL age group is shown by the spread of the age readings errors. It appears
to be no RELATIVE bias, if the age reading errors are normally distributed. The distributions are
skewed, if RELATIVE bias occurs.
References 22
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
Appendix 3: Details results of Trachurus picturatus
The number of age readings, the coefficient of variation (CV), the percentage of agreement and
the RELATIVE bias are presented by MODAL age for each age reader and for all readers
combined. A weighted mean CV and a weighted mean percent agreement are given by reader and
all readers combined. The CV's by MODAL age for each individual age reader and all readers
combined indicate the precision in age reading by MODAL age. The weighted mean CV's over all
MODAL age groups combined indicate the precision in age reading by reader and for all age
readers combined.
References 23
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
References 24
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
In the age bias plots below the mean age recorded +/- 2stdev of each age reader and all readers
combined are plotted against the MODAL age. The estimated mean age corresponds to MODAL
age, if the estimated mean age is on the 1:1 equilibrium line (solid line). RELATIVE bias is the
age difference between estimated mean age and MODAL age.
25
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
The coefficient of variation (CV%), percentage of agreement and the standard deviation (STDEV)
are plotted against MODAL age. CV is much less age dependent than the standard deviation
(STDEV) and the percentage of agreement. CV is therefore a better index for the precision in age
reading. Problems in age reading are indicated by relatively high CV's at age.
The distribution of the age reading errors in percentage by MODAL age as observed from the
whole group of age readers in an age reading comparison to MODAL age. The achieved precision
in age reading by MODAL age group is shown by the spread of the age readings errors. It appears
to be no RELATIVE bias, if the age reading errors are normally distributed. The distributions are
skewed, if RELATIVE bias occurs.
References 26
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
Appendix 4: Details results of Trachurus mediterraneus
The number of age readings, the coefficient of variation (CV), the percentage of agreement and
the RELATIVE bias are presented by MODAL age for each age reader and for all readers
combined. A weighted mean CV and a weighted mean percent agreement are given by reader and
all readers combined. The CV's by MODAL age for each individual age reader and all readers
combined indicate the precision in age reading by MODAL age. The weighted mean CV's over all
MODAL age groups combined indicate the precision in age reading by reader and for all age
readers combined.
References 27
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
References 28
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
In the age bias plots below the mean age recorded +/- 2stdev of each age reader and all readers
combined are plotted against the MODAL age. The estimated mean age corresponds to MODAL
age, if the estimated mean age is on the 1:1 equilibrium line (solid line). RELATIVE bias is the
age difference between estimated mean age and MODAL age.
References 29
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
30
Report of the Horse Mackerel, Mediterranean Horse Mackerel and Blue Jack Mackerel (Trachurus trachurus, T.
mediterreaneus and T. pictatus) Otolith Exchange 2015
The coefficient of variation (CV%), percentage of agreement and the standard deviation (STDEV)
are plotted against MODAL age. CV is much less age dependent than the standard deviation
(STDEV) and the percentage of agreement. CV is therefore a better index for the precision in age
reading. Problems in age reading are indicated by relatively high CV's at age.
The distribution of the age reading errors in percentage by MODAL age as observed from the
whole group of age readers in an age reading comparison to MODAL age. The achieved precision
in age reading by MODAL age group is shown by the spread of the age readings errors. It appears
to be no RELATIVE bias, if the age reading errors are normally distributed. The distributions are
skewed, if RELATIVE bias occurs.