Quick link to: EFIMAS home, WP4, CS6

FL-ISIS approach for the Case Study 6

Description of the fishery, stocks and management system

The Nephrops stock have no defined biological reference points but it is considered to be inside safe biological limits according to the last ices advice (ICES, 2005). However, Nephrops fishery generate a large amount of discards (up to 60 % in number) and since 2004, the use of selective grids on trawls making easier the juvenile escape is mandatory. The Nephrops fishery has a major impact on the Northern Hake stock because the Nephrops fishing grounds are on hake nurseries. Then, discards of young Hake in Nephrops fisheries are very important. The latest assessments made by ICES in the last three years indicated that this Hake stock was outside safe biological limits and an emergency plan has been implemented since 2001 but is now follow up by a recovery plan since 2004. It has been essential to suggest regulations to reduce the catches of small Hake in fisheries taking place in Hake nursery areas. Then, a hake box have been designed in the whole Sub-area VIII with 100 mm of mesh size for all trawlers. A settle of a licence system (numerus clausus) have also been implemented. However, since it does not exist any fleet explicitly targeting Hake alone, the use of uniform regulations to safe Hake is problematic since any management approach affects all fleets having these by-catches.

Data and parameters

The Hake/Nephrops fishery in the Bay of Biscay generates a large amount of by-catch of juvenile hake, especially at recruitment time. The technical interaction between these two stocks, its effect on stock evolutions under different management procedures need to be addressed in the same simulation framework. At the same time, selective devices to reduce important discards occuring in the Nephrops fishery is tested.

see tables of data: parametrization tables

stock data

Drouineau et al. 2006 previously applied the ISIS-Fish model (Mahevas and Pelletier 2004) on the mixed fishery of hake/nephrops of the Bay of Biscay and the present study partly reuses values of the biological operating model parameters. The spatial distribution of Nephrops is restricted to the 'Grande Vasiere' (ICES VIII) whereas Hake is located along all the shelf of the Bay of Biscay and the Celtic sea (ICES VII). The effect of the reduction to the Bay of Biscay area in the present study is not well known and it is assumed that reproduction and recruitment areas may be distinct between this region and the Celtic Sea like ICES which assume two distincts stocks. Initial population abundances were taken from the 2003 ICES stock assessement and were uniformly distributed over the area of presence of the stock in the Bay of Biscay. Nephrops stock is sedentary while Hake stock show seasonal ontogenic migrations between areas with agregation on reproduction zone along the shelf edge at the beginning of the reproduction season (January) and dispersion over the whole of presence area at the end of the season (July); Quero 1997). The migration in or out of the Bay of Biscay have not been taken into account because of lack of data. For both stocks, age and season-specific catchabilities were estimated based on data analysis of log-books data and catch-at-length information (Drouineau et al. 2006).

fleet data

The Hake/Nephrops mixed fishery is located on the continental shelf of the Bay of Biscay (ICES VIIIab) and involved both french and spanish fleets. Fleet definition for this fishery was performed by Mahevas et al. (2006) in the TECTAC project framework. The modelled french fishing activity is a set of 8 FLSetOfVessels i.e. a total of 206 vessels distributed into 2 vessel types (intermediate: 12-24 meters; large: > 24 meters) attached to 1 harbour (Guilvinec) and displaying 4 strategies of activity pattern over the year. The spanish fleet is described by 6 FLSetOfVessels i.e. a total of 69 vessels distributed into 2 vessel types (20-29 meters; 30-39 meters) attached to 2 harbours (Pasaia and Ondarroa) and displaying 2 strategies (hake and Megrim or Anglerfish as main target respectively) with specific trip types attached to these fleets. Spatio-temporal distribution of fleet efforts are driven by related-metiers zones and activities over year using a static allocation of effort per metier over time obtained from data analysis (i.e. a constrained dynamic). Metiers use either single or pair bottom trawls or gillnets for which the age-specific selective behaviour have been assessed (Mahevas et al. 2006). The spanish fleet definition result from analysis of logbook and activity data from AZTI-Tecnalia database.

economic data

Related-economic parameters for french fleets and metiers and market place parameters were obtained from data collected by the Fisheries Information System of IFREMER via surveys of individual vessel owners (Mahevas et al. 2006). Spanish economic data are obtained from the Statistical Service of the Departement of Agriculture (Fisheries and Food of the Basque Government). In addition to the Nephrops and the Hake stocks, the set of targeted species (i.e. species constituting a large part in the fleet revenue) is the Black Anglerfish (Lophius budegassa), Monkfish (Lophius Piscatorius) and Megrim (Lepidorhombus whiffiagonis). These species have not been explicitely modelled but revenues from these species is taking into account as a linear function of the fleet effort (see Equations in Bastardie et al. 2007).


Development of the ISIS-Fish operating model recoded into the FLR

ISIS-FLR model defines a set of classes and methods encapsulated in a R package named ‘FLIsis’ which is available for download on the FLR web site http://flr-project.org/doku.php?id=appl:isisflr. The conceptual representation of fishery of the ‘ISIS-FLR’ model were being built from the ‘ISIS-Fish’ software (Mahevas and Pelletier 2004). This model projects forward initial populations of given stocks impacted by fishing.

Examples of R code to run FLIsis simulation:


# run a simulation (baseline sce. with default settings)
flisis <- FLIsis(namesim='70mmNoMPA' ,  set.paths= list(my.path.output="C:"),
            set.general= list(nbyears=10), set.rules= list(), set.stochast= list(nbiter=1))
flisis <- init(flisis)

# run a simulation (with a management rule e.g. MPA)
flisis <- FLIsis(namesim='70mmMPA' , 
           set.paths= list(my.path.output="C:"),
            set.general= list(nbyears=10),
             set.rules= list(closure.f=TRUE, closure.metiers= "all",closure.areas = c("23E6", "22E6", "22E7", "23E7", "20E8", "19E8"), 
               closure.seasons = c("feb", "mar", "apr", "may", "jun"), closure.years=c(2:10)),
                set.stochast= list(nbiter=1))
flisis <- init(flisis)

# run a simulation (with a management rule e.g. 130mm in mesh size )
flisis <- FLIsis(namesim='130mmNoMPA' , 
           set.paths= list(my.path.output="C:"),
            set.general= list(nbyears=10),
             set.rules= list(selectivity.f=TRUE, selectivity.values=c(130), selectivity.gears=c("OTBLN","TTBLN")),
                set.stochast= list(nbiter=1))
flisis <- init(flisis)

# run a simulation (with a combination of management rules e.g. 130mm in mesh size / MPA)
flisis <- FLIsis(namesim='130mmMPA' , 
           set.paths= list(my.path.output="C:"),
            set.general= list(nbyears=10),
             set.rules= list(selectivity.f=TRUE, selectivity.values=c(130), selectivity.gears=c("OTBLN","TTBLN"),
              closure.f=TRUE, closure.metiers= "all",closure.areas = c("23E6", "22E6", "22E7", "23E7", "20E8", "19E8"), 
               closure.seasons = c("feb", "mar", "apr", "may", "jun"), closure.years=c(2:10)),
                set.stochast= list(nbiter=1))
flisis <- init(flisis)

## general plot from FLIsis
        per.month.catches=TRUE, per.month.abundance=FALSE,

## mapping on the fishery region

## example for plotting per fleet-segment
FrenchTrawlers <- c("BenthicIntermediate","BenthicLarge","HakeIntermediate","HakeLarge","NephropsIntermediate","NephropsLarge")
FrenchNetters <- names(fleets) [grep("Net",names(fleets))]
SpanishTrawlers <- c(names(fleets) [grep("2029",names(fleets))], names(fleets) [grep("3039",names(fleets))])

#  IN the mpa
per.season <- TRUE
if(per.season==FALSE) ylimits <- list(nephrops=c(0,0.4),hake=c(0,4))
if(per.season==TRUE)  ylimits <- list(nephrops=c(0,0.2),hake=c(0,0.20))
par(mar=c(4,4,0.3,0.3)) # margin
par(mgp=c(2, 0.7, 0))  # margin line (labels)
for (spp in flisis@inputdata$STOCK.NAMES){
  for(typedata in c("Landings","Discards")){
  # 1. landings and discards in MPA region - MPA sce/no MPA sce
  plots (fleets.1, type.data= typedata, spp= spp, set.of.fleet.names= list(FrenchTrawlers,FrenchNetters,SpanishTrawlers),
        all.fleet.names= names(fleets.1), ylimits=ylimits[[spp]], leg.labels= c("FrenchTrawlers","FrenchNetters","SpanishTrawlers"),
         a.region= flisis@rules$closure.areas, exclude.region=FALSE, per.season= per.season)
  plots (fleets.2, type.data= typedata, spp= spp, set.of.fleet.names= list(FrenchTrawlers,FrenchNetters,SpanishTrawlers),
        all.fleet.names= names(fleets.2), ylimits=ylimits[[spp]], leg.labels= c("FrenchTrawlers","FrenchNetters","SpanishTrawlers"),
         a.region= flisis@rules$closure.areas, exclude.region=FALSE, add=TRUE, per.season= per.season)

#  OUT the mpa 
# => same as previously but setting exclude.region to TRUE

The model is a spatially and seasonally explicit multi-fleet, multi-stock model taking into account of the response of fishermen to a range of management rules through dynamic allocation of fishing effort. Fishery simulations result from the interaction between four sub-models : (i) a population sub-model which simulate the demographic processes for each species over space and time; linked with (ii) a exploitation sub-model by the relation between fishing effort of the fleets and fishing mortality; the fishing mortality depending on (iii) the management sub-model altering the spatio-temporal distribution of the fishing effort from the response of fisher behaviours to management constraints; finally (iv) a bioeconomic model compute revenue from landings, fish price and costs.

Beyond traditional management rules (e.g. T.A.C.) it is assumed that alternative management rules could be suggested by monitoring the spatio-temporal covariance between population and exploitation dynamics. Then, at each time step, fisheries dynamics is determined by the spatial overlap between population areas, fishing activity areas and management areas. The model is a monthly time step model and uses a regular grid of cells to discretize the fishery region (the ices square resolution of space by default) and both these features could be set up as needed by the user.

Figure 2: Representation of the ISIS-FLR fisheries management model as a set of linked FLR-shaped classes into 4 sub-models (I: management rules model; II: biological model; III: fleet activities model; IV: economic model). The internal structure of classes (i.e. slots e.g. 'n' for 'FLBiol2') are not described for all classes and, for described classes, only main slots are listed on this picture. dotted line: links between classes during the simulation process.

The goal of recoding the ISIS-Fish operating model using FLR (i.e. biological, fleet and economic OM) participate of the willing to plug in this existing model to the FLR platform. Hence, the set of tools and packages developed in FLR could be reused to implement blocks of a management loop lacking in ISIS-Fish software such as observation error model, assessement model, etc. The translation of spatially- and seasonnally- explicit ISIS-Fish objects toward FLR environment have been facilited using FLR existing classes and in particular the FLR lower-level class named 'FLQuant' which provide the basic structure to design time and area-desagregated arrays of data. 'FLQuant' objects are 5-dimensions array of data: the first dimension can be set by the users (e.g. 'Age', 'Length', 'Vessel type') and the other dimensions are 'Year', 'Unit' (e.g. 'Sex', 'Spawning type'), 'Season' and 'Area'. These dimensions are common with the existing ISIS-Fish's basic object dimensions. Different FLQuants have been grouped into FLR classes and constitute attributes of these classes (generally called 'slots' in the R OOP vocabular; Chambers 2000).

Figure 2 depicts the representation of the fishery of the present model as a set of composite FLR-shaped classes nested by common slots acting as relational keys between objects. For the sake of simplicity, related-fishery region data used as input in ISIS-FLR are currently stored as text files and consequently do not require additional software than R statistical language. Data are loaded in the R working space and constitute a list of slots in the main class of the ISIS-FLR application we called 'FLIsis'. FLIsis acts as a class manager by holding the paths for sourcing and computation, the set of data, the name of simulations and the combination of management rules (and related parameters) to be tested by sequential simulations.

Then, all ISIS-FLR objects are initially created at the beginning of the simulation and act as collectors of the data processed by the simulation loop and equation computation over time steps (Table 1).

population dynamic submodel

Recent models in fisheries sciences aims at taking into account of space and time distribution of the abundance of the populations and their demographic processes possibly varying in the course of life cycle of species. Further, since some demographic processes have been shown to be age-specific (e.g. migration, mortality) the population dynamic is simulated in a age-structured way. In ISIS-FLR, successive demographic processes could occur depending on relevant areas and time steps:

  • class change. Due to ageing, classe change occurs either the 1st of January if the population is age-structured, or at the beginning of all time steps if a length-structured population is simulated. In this last case, the proportion of individual changing class from a step to the next is predicted from the common reversal Von Bertalanffy relationship.
  • seasonal large scale ontogenic migration. Possible age-specific migrations are simulated using coefficients which are specifically applied on the spatially distributed abundance as a proportion of migrants from an area to another.
  • reproduction. Egg production occurs in reproduction cells from age desagregated spawner abundance and fecundity-at-age in timing given by the time desagregated ogive of reproduction. From a step to the next a egg mortality is applied on the pool of egg and larva.
  • recruitment. New recruited from the larval pool are uniformly distributed between recruitment cells in function to the timing given by the ogive of recruitment. Hence, the stock-recruitment relationship is split up into a stock-eggs relationship and the application of an eggs natural mortality.

Simulating dynamic of populations is facilited by the use of the area and season-desaggregated FLQuant objects. Hence, the class 'FLBiol2'; extended from FLBiol) defines a set of FLQuant objects which aims at holding all the age-, time- and area-desaggregated relevant information to compute demographic processes (fecundity, natural mortality, etc.). A FLBiol2 object is then defined over season and cells which constitute the area of distribution of the studied stock for all time steps. The information about recruitment cells, reproduction cells or other type of zones constitute slots of FLBiol2 class, as well as a slot concerning migration coefficients from zones to another, and a slot storing a matrix of 'age' (or 'length' or 'stage') change.

FLMetier: a FLR-shaped class to take into account of spatio-temporal heterogeneity of fishing effort for a same fleet

Several fleets are supported as set of vessels with distinct physical characteristics (e.g. length, engine power, etc.) mainly because they result in distinct travelling times, fishing powers and corresponding costs. Further, to take into account of spatio-temporal heterogeneity of fishing activities, 'metiers' (also called 'trip type', 'rigging' or only 'gear') are defined to constitute each a particular spatio-temporal fishing activity of the fleet.

A metier is defined as a unique arrangement of a gear, one or a set of target species and a fishing ground. Within each strategy, the distribution of fishing effort between metiers over the year is dictated either by a static activity calendar per metier (a particular type of calendar constitute a strategy) or by a dynamic allocation of effort model into metiers (e.g. given by the outputs of a Random Utility Model, etc.) and mimics the splitting up of the global effort of the strategy into effort per metier.

Hence, ISIS-FLR totally reuses the spatio-temporal fleet effort desagregation into metiers described in ISIS-Fish. To perform this, the development of new designs of FLR-shaped classes (Table 1) were required to store specific features of the ISIS-Fish way of modelling. In particular, FLSetOfVessels (an extended version of the common FLFleet class) enables to define a set of vessels sharing physical characteristics but also sharing the same strategy i.e. the same sequence of fishing activities or metiers over time. Consequently, each fleet could be modelled by one or several FLSetOfVessels objects, as many as possible strategies. Each FLSetOfVessels object has got a list of FLMetier objects to describe the metiers practiced within a given strategy. The respective part of activity for each metier of the FLSetOfVessels object is hold in its particular slot named 'activity' having metier names as first dimension. Then, various fishing activities occuring on different areas and possibly varying over time could be set up for each FLsetOfVessels. Metier zones constitute a list of cells among fishery region cells.

Economic submodel: FLRevSOV, FLRevMetier, FLCostSOV, FLCostMetier, FLMarket

The economic sub-model explicits fleet and metier cost definition and dynamic of fleet revenues from landings. The aim is to (i) assess the short-term economic effects and viability of management measures, and (ii) take into account the short-term response of fishers (i.e. in terms of spatio-temporal reallocation of effort) to economic indicators or market events. Then, owner margin and vessel margin are computed from economic characteristics of metiers as detailed in Bastardie et al. (2007).

Metier-stock specific prices depends on total landings for each stock on the market place attached to a given metier. Gross return for a particular metier are computed from fish price and age-structured landings of the metier. The gross return from other species (i.e. species whose population dynamics is not explicitaly simulated) is added as a linear function of metier effort. Net revenues are then computed from gross return minus costs for fishing. The cost structure of the fishing activity is both fleet- (insurance costs, etc.) and metier-specific (fuel costs may depends on the used gear and the visited area, etc.). It is assumed that all vessels of a given fleet share similar cost structure which seems realistic since fleets are set of vessels with similar physical characteristics. Cost structure is constants over the simulation process while the total cost depends on the effort allocated on each fishing activity at eac time step. The last version of ISIS-Fish have recently added an economic box (version 2.0).

Additional new FLR-shaped classes were required to take into account of economic terms named FLMarket, FLCostSOV, FLCostMetier, FLRevSOV, FLRevMetier (Table 1) which respectively impact dynamics at either fleet level or at metier level. Objects of these classes are linked with FLSetOfVessels and FLMetier objects using respectively the name of the set of vessels and the name of the metier as relational keys. Then, a list of FLMarket objects stores stock-specific parameters attached to a market place to compute age-structured fish price from a price equation (e.g. supply and offer law) using the total landings on the concerned market. Capital dynamics is also supported through these classes and may be used to implement an entry/exit model of vessels to/from the fishery. Equally, a model option enables economic outputs to influence the dynamic of effort allocation between metiers for the next time step.

Management rules

The settings for management sub-model are stored into a new FLR-shaped class named 'FLSetOfRules'. Management rules runs over time steps either at predefined steps from the database or dynamically in response to simulation events (e.g. TAC reached, etc.). Hence, a combination of management rules and all their related values can be tested by sending this combination as argument during the creation of the FLIsis manager class. Using a grid of simulations, these management rules are evaluated before each new simulation using new values as arguments of each rules. Conversely, management rules could occur over the simulation depending on internal events.

A particular management rule implemented in ISIS-FLR is MPA rule. With this rule, Marine Protected Areas could be set up as a partial or total closure of all or chosen fishing activities on predefined or dynamic zones at predefined or dynamic times. Because the model is spatially explicit with a monthly time step, we may consider management measures that apply either during some months throughout the year, and are either global (at the scale of the region) or local (within a particular zone). In this way, a large variety of MPA designs may be evaluated. Assumptions are needed concerning fishers behaviours facing to the management rule constraint. In case of MPAs, it is assumed that the activities of possible metier for each fleet are reallocated to the remaining cells of metier zones. If all metier cells are closed for a given metier, the activity of the concerned metier is reallocated to other possible metiers of the fleet. Finally, if all metier cells are closed for all possible metiers, the activity of the concerned fleet is set up to 0.

The alteration of the selectivity of the used gears could also be set up. If this rule occurs, selectivity curves from gear-specific selectivity equations are re-calculated using desired values for the technical parameter (e.g. mesh size).

Table 1: FLR-shaped classes of the ISIS-FLR model

name origin description
Manager classFLIsisISIS-FLR The manager class of the ISIS-FLR model to compute simulation steps using three methods (1. hold data from a data set and create ISIS-FLR objects, 2. run a set of simulations, 3. run a Graphical User Interface to plot and draw results from a set of simulation)
Basic classFLQuantFLR team lower-level FLR class to create objects which are 5- or 6-dimensional arrays of data i.e. age or length class, year, sex, season, area (,iteration). (See Teach yourself: FLQuants)
Population-related classesFLBiol2inherited from FLBiolClass of objects including a set of FLQuant objects for storing related-population data such as weight, maturity, etc. but also data frames for migrations and class changes. This class is useful to run an operating model describing population dynamics. (see introduction_to_composite_objects)
FLStockFLR teamClass of objects used for storing the observed population dynamics of a fish stock in the model (catches, landings, etc.)
Exploitation-related classesFLCatch inherited from FLCatchmodified from FLCatchClass of objects including a set of FLQuant objects storing stock-specific parameters (catchability, etc.) and catches, discards and landings
FLMetierextended from FLMetierClass of objects aiming at storing the part of a a fleet activity corresponding to a unique arrangement of target species, gear and fishing zone
FLSetOfVesselsextended from FLFleetClass of objects used to describe a part of the activity of a fleet i.e. sets of vessels sharing physical characteristics and practising possible metiers following a given strategy. A strategy is defined as a same sequence of metiers over the year with partition of the fleet activity split up into metiers. Then, each fleet in the model is represented by one or several FLSetOfVessels objects (as many as possible strategies. Each FLSetOfVessels object includes a list of FLMetier objects (as many as possible fishing activities displaying by the fleet). Equally, a FLMetier object includes a list of FLCatch objects (as many as stocks which can be fished by the metier)
FLGearISIS-FLRClass of objects storing the gear properties (selectivity equation, standard factor of the gear, etc.)
FLDiscardsISIS-FLRClass of objects storing the discard behaviour (ogive of discards, etc.)
Management-related classesFLMgtRulesISIS-FLR storing various settings for management rules.
Economic classesFLCostSOVISIS-FLRstoring various input economic costs at set of vessels level.
FLCostMetierISIS-FLR storing various input economic costs at the metier level.
FLRevSOVISIS-FLRstoring various output economic variables at the set of vessels level
FLRevMetierISIS-FLRstoring various output economic variables at the metier level
FLMarketISIS-FLR storing price equation parameters attached to a particular market place to compute variable fish price

Conditionning OM (stock, fleet and economic)

only projections forward of abundance and catches are performed with ISIS-FLR.

  • stock: parameter values (M, growth, spatio-temporal distribution, fecundity, maturity) result from the literature, Initial population numbers from the WGHMM 2003.
  • fleet: no conditionning of effort from historical catch-at-age matrix: effort per fleet per metier is explicitly modelled as a number of trips per month, times a time for fishing, times the proportion of the fleet activity into the metier, times the number of vessels in the fleet. By default, the proportion of fleet activity into a given metier is a constrained dynamic from calendar activity data analysis.

observation error model and assessement model

FLIsis aims not to produce by himself a complete management procedure but rather provides a deterministic operating model directly impacted by various management rules. Hence, no added errors on abundances and catch-at-age are taking into account inside FLIsis. However, a run of sequential simulations using a range of values for input parameters could lead to compute elasticity coefficients. Further, FLIsis outputs should act as inputs for other FLR packages and in particular the FLOEM package dedicated to the observation error model.

Management scenarios

Deterministic simulations were running to test two management scenarios from the data set which defined the initial conditions assuming total compliance of fishermen for both scenarios:

  • MPA scenario. Northern hake use to spawn from February through July along the shelf edge, the main areas extending from north of the Bay of Biscay to the south and west Ireland (ICES 2006). Large part of the hake catches are obtained during the first quarter of the year when the spawning period occurs. Data have shown that there are specific ICES rectangles where high catches occur reiteratively year after year, and in particular for the spanish fleets using VHVO gears (i.e. 'Very High Vertical Opening' nets). Thus, a possible management scenario is to close these rectangles (contiguous 22E4, 23E5, 23E4 and 22E5; contiguous 20E7 and 19E7) to the VHVO pair trawlers from February to May.
  • selective gear scenario. In order to reduce the Hake discards from by-catch of juveniles for the fleets targeting nephrops the impact on an increasing minimum mesh size for trawl gears (70 mm, 100 mm, 130 mm which respectively correspond to the 50 percent retention age in years of 2.5, 4.4 and 6.3 for the Hake and 0.5, 1.2 and 2.1 for Nephrops with our settings for trawl gears) is tested on the fishery in terms of stock evolution, landings and fleet revenues.

The combination of both the tested regulations should be able to identify cross interaction between their respective effects.


ISIS-FLR operating model

A main issue in the dynamics of mixed fisheries is that of technical interactions, leading to incidental catch and discarding. Technical interactions largely depend on the allocation of fishing effort between métiers and fishing grounds which in turn is tightly linked to economic conditions and to the expected profitability of alternative options for fishing effort allocation. We developed a bioeconomic fishery model to investigate these issues, and in particular to explore the possibilities of mitigating these interactions through appropriate policy options. The bioeconomic model was developed using and creating FLR (http://www.flr-project.org) packages as part of the EFIMAS project http://europa.eu.int/comm/research/fp6/ssp/efimas_en.htm). The model is spatially- and seasonally- explicit, it considers population dynamics, exploitation dynamics and policies are explicitly modelled, building in the fishery model underlying the ISIS-Fish software (http://www.ifremer.fr/isis-fish\). The model is applied to the hake-nephrops fishery in the Bay of Biscay. The fishery generates a large amount of by-catch of juvenile hake, particularly at recruitment time. We modelled the dynamics of the main fleets exploiting hake and nephrops, and investigated the consequences of alternative policy options, including Marine Protected Areas, closed seasons, and selective devices.

Hake/Nephrops mixed fishery

submitted paper

Figure - Simulated a. Abundance, b. SSB, c. Landings, d. Fbar for Hake for under the different tested scenarios.

Figure - Simulated a. Abundance, b. SSB, c. Landings, d. Fbar for Nephrops for under the different tested scenarios.

Figure - Hake and Nephrops landings and discards over all areas for the main fleet-segments, 70mm/MPA sce. in dotted line, 70mm/noMPA sce. (i.e. the baseline scenario) in solid line

Figure - Hake and Nephrops landings and discards for the main fleet-segments INSIDE the MPA zone, 70mm/MPA sce. in dotted line, 70mm/noMPA sce. (i.e. the baseline scenario) in solid line

Figure - Hake and Nephrops landings and discards for the main fleet-segments OUTSIDE the MPA zone, 70mm/MPA sce. in dotted line, 70mm/noMPA sce. (i.e. the baseline scenario) in solid line

Figure - Selectivity effect on Hake and Nephrops landings and discards for the main fleet-segments, 130mm/noMPA sce. in dotted line, 70mm/noMPA sce. (i.e. the baseline scenario) in solid line

Figure - Conjugued effect of selectivity measure and MPA regulation on landings and discards for the main fleet-segments, 130mm/MPA sce. in dotted line, 70mm/noMPA sce. (i.e. the baseline scenario) in solid line


Bastardie F., Pelletier D., Mahevas S., Guyader O., Thebaud O., Sauturtun M., Prellezo R. 2008 ISIS-FLR: An FLR-based bioeconomic operating model for mixed fisheries: framework and application submitted

submitted paper


Bastardie F., Pelletier D., Mahevas S., Guyader O., Thebaud O., Sauturtun M., Prellezo R. 2008 ISIS-FLR: An FLR-based bioeconomic operating model for mixed fisheries: framework and application submitted

Chambers, J., M. 2000. Programming with data - a guide of S language, Mathsoft.

Drouineau, H., Mahevas, S., Pelletier, D., Beliaeff, B. 2006. Assessing the impact of different management options using ISIS-Fish: the French Merluccius merluccius - Nephrops norvegicus mixed fishery of the Bay of Biscay. Aquat. Living Resour., 19, 15-29

Mahevas S., Pelletier D. 2004. ISIS-Fish, a generic and spatially explicit simulation tool for evaluating the impact of management measures on fisheries dynamics. Ecological Modelling, 171, 65-84

Mahevas, S., Bertignac, M., Daures, F., Guyader, O., Marchal, P., Pelletier, D., Prellezo, R., Santurtun, M., Thebaud, O. 2006. Hake/Nephrops mixed fishery of the Bay of Biscay: ISIS-Fish 2.0 parametrization in TECTAC report.

Fr. Bastardie 2007/04/16 13:39

efimas1/wp4/cs6/appr2/main.txt · Last modified: 2008/07/16 13:06 by fb
Except where otherwise noted, content on this wiki is licensed under the following license:CC Attribution-Noncommercial-Share Alike 3.0 Unported
Recent changes RSS feed Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki