Docstoc

WD_catchabilitySurveys_anchovy

Document Sample
WD_catchabilitySurveys_anchovy Powered By Docstoc
					ICES / CIEM
International Council for the                                                WGACEGG
Exploration of the Sea                                                       Lisbon, 27/11 – 01/12, 2006

Working document



                    Catchability analysis between abundance estimates
                           from DEPM and Acoustic methods

                     Leire Ibaibarriaga (AZTI) and Pierre Petitgas (IFREMER)


1. Introduction

The bay of Biscay anchovy is assessed every year using two independent research surveys. On the one
hand, the acoustic survey conducted by IFREMER, and on the other hand, the survey for the
implementation of the DEPM (Daily Egg Production Method) carried out by AZTI. Both surveys occur at
the main spawning period of the species (May-June) and attempt to cover its whole spatial distribution.
However, in many years both biomass estimates disagree and provide two different perceptions of the
state of the stock.

This document attempts to analyse the relative catchability between the acoustic and the DEPM biomass
estimates from 1994 to 2005. Firstly, the data sources are described. Then, the catchability analysis is
conducted following two approaches. The first one consists on analysing the time-series of the overall
biomass estimates. The second one, a more in-depth exploration, tries to compare acoustic biomass and
daily egg production estimates in space. Finally, the main results are discussed and conclusions to feed in
posterior Work Packages are extracted.


2. Material and Methods

The acoustic surveys are conducted since 1987, their main objective being to study the abundance and
distribution of pelagic fish in the Bay of Biscay. This method consists on mapping the area of interest
using echo-sounders, allocating the energy to each of the species with the help of evenly distributed
fishing hauls and transforming this energy into biomass. Thus, this method provides spatially explicit
biomass estimates for the Bay of Biscay anchovy stock.

The DEPM (Parker 1980, Lasker 1985) consists on estimating the spawning stock biomass as the ratio
between the daily egg production rate estimated from an ichthyoplanckton survey conducted during the
peak of the spawning period and the daily specific fecundity which is estimated from a parallel adult
sampling. Since 1987 this method has been routinely applied to estimate the bay of Biscay anchovy
biomass (REF), with the exception of 1993 in which no survey was conducted. Even if the DEPM
sampling strategy is spatially explicit (the survey covers the whole spawning area and each sample is
associated to a location), in the traditional method all the parameters involved in the calculation of the
biomass are assumed to be constant in space. However, in a recent EU project, a new methodology using
generalized additive models (GAMs) allowed to obtain daily egg production estimates in space. This
methodology was applied to the time series of DEPM surveys, by fitting GAMs with overdispersed
Poisson distribution and log link to the number of eggs found per day. The mortality rate was assumed
constant over the space and the daily egg production rate was modelled by two bivariate smooth terms:
one representing the spatial location, s(Long, Lat), and one representing the environmental conditions,
s(Sst.fit, Sss.fit). The biomass estimates in space were derived by assuming that daily fecundity was
constant in space. Daily fecundity can also be estimated in space by using GAMs. However, as the major
source of the spatial variability in biomass is due to the egg distribution, this approach was considered
good enough.

For the first part of the analysis, the time series of total biomass estimates from Acoustics and DEPM that
are yearly reported to the ICES Working Group on assessment on Mackerel, Horse mackerel, Sardine and



                                                    1
Anchovy (WGMHSA) are analysed. This WG provides synthetic estimates of the state of the stock by
combining all the available information, both research surveys (Acoustics and DEPM), and the fishery
data, into an integral assessment model. In particular, the bay of Biscay anchovy stock is assessed using
the Integrated Catch-at-Age (ICA) (Patterson and Melvin 1996). The biomass estimates from the
Acoustics and DEPM together with the ICA results (Figure 1) were taken from the last ICES WG (ICES
2006).

The second part of the analysis, namely, the spatial comparison between the Acoustic and DEPM
biomass, was conducted using spatially disaggregated biomass estimates from Acoustics and the gam-
based biomass estimates from DEPM. Figure 2 shows the timing of the acoustic and DEPM surveys from
1989 to 2005. Up to 1992 the acoustic survey used to take place in April whereas the DEPM survey was
on May-June. Only data from 1994 onwards, when both surveys are coincident in time, are selected for
this part of the analysis. In addition, given that the area covered in both surveys is not exactly the same,
the analysis was focused on an agreed area common to both surveys and all the years (Figure 3). A
regular grid was defined in this area. Five different strata (Figure 3) were defined within the area:

        Strata 1: Below 44º
        Strata 2: Between 44º and 45º and below 200m depth
        Strata 3: Between 44º and 45º and above 200m depth
        Strata 4: Above 45º and below 200m depth
        Strata 5: Above 45º and above 200m depth


3. Time-series analysis

The q statistic between two abundance indices I1 and I2 is defined as a ratio of one survey index to the
sum of the two indices to caracterise the difference in relative catchability across the years:

                                                       I1
                                               q            .
                                                    I1  I 2

This statistic takes values between 0 and 1. Values above 0.5 indicate that I 1 is larger than I2, and values
below 0.5 otherwise. Moreover, it is a function of the ratio between the catchability parameters of two
abundance indices, i.e., being I1 = q1 A and I2 = q2 A (assuming no errors) where q1 and q2 are the
catchability coefficients of the I1 and I2 indices of the real abundance A, it follows that

                                                       q1
                                             q1 A         q2
                                      q            
                                         q1 A  q2 A q1  1
                                                        q2

The q statistic was computed for the acoustics (as I1) and DEPM (as I2) biomass estimates reported to the
WGMHSA (ICES 2005). Since 2000 acoustics biomass estimates are larger than DEPM ones, with an
average relative catchability of 0.65 with standard deviation of 0.9 (Figure 4). The autocorrelation
function of the q statistic (Figure 4) does not show any significant pattern.

Scatter plots between the relative catchability and some of the geostatistical statistics from the egg
distribution and the numbers at age 1 estimated by ICA are shown in Figure 5. The relative catchability is
positively correlated with anisotropy in the spatial distribution of the eggs. In the recent years, the
acoustic index (I1) has been systematically above the DEPM index (I2), which could be explained by a a
change in the spatial distributions.


4. Comparison in space

Spatial egg abundance, gam-based DEPM biomass and acoustic biomass estimates from 1994 to 2005 are
shown in Figures 6 a, b, c, d, e.




                                                     2
At each point of the common grid, s, the q statistic between two abundance indices I1 and I2 is defined as

                                                             I1 ( s)
                                            q( s )                       .
                                                       I1 ( s)  I 2 ( s)

The maps of the q statistic in space are shown in Figure 7.

The center of gravity and inertia between egg abundance, DEPM biomass and acoustic biomass are
shown in Figure 8. In 1994 and 2003 the three distributions were very close to each other. In the rest of
the years, the acoustic biomass distribution has the center of gravity closer to the coast, which might be
due to the drift of the eggs. The axis of the inertia of the three distributions is similar in all the years
except in 1998.

The center of gravity and inertia of the q statistic between acoustics and DEPM biomass estimates are
similar in all the years (Figure 9). The inertia first axis is parallel to the shelf break, except in 1994 that is
parallel to the coastline, and the second axis is perpendicular to the coastline.

The analysis of the patchiness of the q statistic (Figure 10) shows that there are two types of distributions:
1997 and 2001 in which two patches are distinguished, one in the southern area and one in the northern
area, and the rest of years for which 3 different patches can be found: one in the southern area, below
44.5ºN or 45ºN, one in the middle area and one in the northern area close to the Garonne river mouth.

The comparison between the acoustic and DEPM biomass estimates in the area selected and by strata is
shown in Figure 11. The q statistic in the area selected and by strata is shown in Figure 12.

5. Conclusions

The acoustic and DEPM abundance indices were compared using a constructed statistic q that
characterised the relative catchability between surveys. The statistic q was computed globally on the
survey indices. The spatial distributions in the acoustic and DEPM abundance were compared using
spatial indicators develop within the EU project FISBOAT to explain differences in q. The largest
differences in q (Fig. 4) that could be interpreted spatially (1998, 2000, 2002, 2004, 2005) were
concomittent with the largest differences in the gravity centres of the spatial distributions of the acoustic
and DEPM abundance (Fig.8). In recent years, the spatial distribution of the adult anchovy and their eggs
has been more coastal and more anisotropic and this could have affected differently the availability to
each survey. A combined index is perhaps more reliable than each index alone.


6. Ackowledgements
This study was partly financed by the UE project FISBOAT (FP6 STREP n°502572).


7. References
ICES 2006. Working Group on assessment on mackerel, horse mackerel, sardine and anchovy.

Patterson and Melvin. 1996. ICA




                                                          3
        140000     acoustics
                   depm
                   ICA
        100000
 SSB
        60000
        20000




                 1990                    1995                       2000                        2005
                                                   Year
Figure 1: Time series of SSB provided by acoustics and DEPM, in comparison to the synthetic estimates
given by the ICA.


                        APRIL                             MAY                               JUNE
         AC
 1989
        EGG
         AC
 1990
        EGG
         AC
 1991
        EGG
         AC
 1992
        EGG
         AC
 1993
        EGG
         AC
 1994
        EGG
         AC
 1995
        EGG
         AC
 1996
        EGG
         AC
 1997
        EGG
         AC
 1998
        EGG
         AC
 1999
        EGG
         AC
 2000
        EGG
         AC
 2001
        EGG
         AC
 2002
        EGG
         AC
 2003
        EGG
         AC
 2004
        EGG
         AC
 2005
        EGG



Figure 2: Timing of the acoustic (light grey) and DEPM (dark grey) surveys from 1989 to 2005.




                                                  4
      47
      46
      45
      44
      43




            -6           -5          -4          -3                   -2       -1



Figure 3: Common area and grid for the spatial comparison between acoustic and DEPM biomass
estimates.
     1.0




                                                               1.0
     0.8




                                                               0.5
     0.6




                                                         ACF
 q
     0.4




                                                               0.0
     0.2
     0.0




                                                               -0.5




           1990       1995        2000        2005                    0    2   4     6    8   10   12
                          Year                                                      Lag

Figure 4: Time series (on the left) and autocorrelation function (on the right) of relative catchability
indices between Acoustics and DEPM biomass estimates reported to the WGMHSA.




                                                     5
    1.0




                                                              1.0
    0.8




                                                              0.8
    0.6




                                                              0.6
q




                                                          q
    0.4




                                                              0.4
    0.2




                                                              0.2
    0.0




                                                              0.0
              5000          10000        15000    20000             1.5       2.0          2.5      3.0          3.5
                              PosArea                                                Anisotropy
    1.0




                                                              1.0
    0.8




                                                              0.8
    0.6




                                                              0.6
q




                                                          q
    0.4




                                                              0.4
    0.2




                                                              0.2
    0.0




                                                              0.0




          0          2000      4000       6000   8000                 14.0   14.5   15.0   15.5   16.0    16.5   17.0
                              N1 (ICA)                                                      Sst

Figure 5: From top to bottom and from left to right, scatter plots between the relative catchability between
the acoustic and DEPM biomass estimates with respect to Positive area from the egg distribution,
Anisotropy from the egg distribution, Numbers at age 1 from ICA and Sea Surface Temperature (SST)
from egg surveys.




                                                          6
                    year 1994 ssbac                                    year 1994 ssbegg                                    year 1994 egg

 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43
      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0


                    year 1997 ssbac                                    year 1997 ssbegg                                    year 1997 egg
 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43




      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0




Figure 6a: From left to right spatial estimates of SSB from acoustics (tonnes/nm2), SSB from DEPM
(tonnes/nm2) and egg abundance (eggs/m2) for 1994 (top panel) and 1997 (bottom panel).




                                                                             7
                    year 1998 ssbac                                    year 1998 ssbegg                                    year 1998 egg

 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43
      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0


                    year 2000 ssbac                                    year 2000 ssbegg                                    year 2000 egg
 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43




      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0




Figure 6b: From left to right spatial estimates of SSB from acoustics (tonnes/nm 2), SSB from DEPM
(tonnes/nm2) and egg abundance (eggs/m2) for 1998 (top panel) and 2000 (bottom panel).




                                                                             8
                    year 2001 ssbac                                    year 2001 ssbegg                                    year 2001 egg

 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43
      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0


                    year 2002 ssbac                                    year 2002 ssbegg                                    year 2002 egg
 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43




      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0




Figure 6c: From left to right spatial estimates of SSB from acoustics (tonnes/nm 2), SSB from DEPM
(tonnes/nm2) and egg abundance (eggs/m2) for 2001 (top panel) and 2002 (bottom panel).




                                                                             9
                    year 2003 ssbac                                    year 2003 ssbegg                                    year 2003 egg

 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43
      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0


                    year 2004 ssbac                                    year 2004 ssbegg                                    year 2004 egg
 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43




      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0




Figure 6d: From left to right spatial estimates of SSB from acoustics (tonnes/nm 2), SSB from DEPM
(tonnes/nm2) and egg abundance (eggs/m2) for 2003 (top panel) and 2004 (bottom panel).




                                                                            10
                    year 2005 ssbac                                    year 2005 ssbegg                                    year 2005 egg

 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43
      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0




Figure 6e: From left to right spatial estimates of SSB from acoustics (tonnes/nm2), SSB from DEPM
(tonnes/nm2) and egg abundance (eggs/m2) for 2005.

                      year 1994 q                                        year 1997 q                                        year 1998 q
 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43




      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0


                      year 2000 q                                        year 2001 q                                        year 2002 q
 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43




      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0


                      year 2003 q                                        year 2004 q                                        year 2005 q
 47




                                                    47




                                                                                                        47
 46




                                                    46




                                                                                                        46
 45




                                                    45




                                                                                                        45
 44




                                                    44




                                                                                                        44
 43




                                                    43




                                                                                                        43




      -3.0   -2.5         -2.0        -1.5   -1.0        -3.0   -2.5         -2.0         -1.5   -1.0        -3.0   -2.5        -2.0       -1.5   -1.0




Figure 7: Spatial distribution of the relative catchability parameter between the SSB estimates from
Acoutics and DEPM.




                                                                            11
Figure 8: Center of gravity and inertia of biomass from acoustics (blue), biomass from DEPM (red) and
egg abundance (pink).


Figure 9: Center of gravity and inertia of the relative catchability between acoustics and DEPM biomass
estimates.




                                                  12
Figure 10: Patchiness of the relative catchability between acoustics and DEPM biomass estimates.


Figure 11: Comparison of acoustics (in black) and DEPM (in red) biomass estimates in the selected area.




Figure 12: Relative catchability of acoustics and DEPM biomass estimates in the selected area.




                                                   13

				
DOCUMENT INFO
Shared By:
Categories:
Stats:
views:6
posted:4/27/2011
language:English
pages:13