Selective fishing gear

A review on the effects of selective fishing gear on cod in the Baltic Sea

Författare: Karl Johan Modig

Handledare: Per Larsson

Termin: VT13

Ämne: Biology

Nivå: Kandidat

Kurskod:2Bi01E

- 1 -

Abstract

The populations of Atlantic cod (Gadus morhua) in the Baltic Sea has been heavily

exploited for decades, with fishing mortalities close to, and for several occasions, even

above one. The larger part of the spawning stock biomass is consequently being

removed each year. The issue of fisheries induced evolution (FIE) has been gaining

attention from researchers lately. The selection pressure driving this evolution is

powered by a connection between high mortality rates and heritable traits. The fishing

in the Baltic Sea is mainly performed with size selective gear that can impose selection

on traits like size-at-age or size-at-maturity. In this review I show how FIE may affect

the Baltic cod towards decreased size-at-age/maturity and how size selective fishing on

stocks at low levels can increase the inherent instability of the population as well as

deprave the Baltic Sea of ecosystemic services from cod.

Keywords Baltic Sea, Atlantic cod (Gadus morhua), fishing mortality, and fisheries induced

evolution

- 2 -

Contents

Introduction _______________________________________________________ - 3 -

The recent history of the Baltic Sea ecosystem ___________________________ - 4 -

Selective fishing _____________________________________________________ - 7 -

Fishery induced mortality ____________________________________________ - 9 -

The rate of genetic change ___________________________________________ - 11 -

Other secondary effects _____________________________________________ - 12 -

Reversing the evolutionary effects of selective fishing ____________________ - 16 -

Discussion ________________________________________________________ - 17 -

References ________________________________________________________ - 19 -

- 3 -

Introduction

The idea that overfishing can lessen future catches, even from the vast seas, can be

found in articles written as early as in the 1800s (Cleghorn, 1855). Some fifty years

later, a theory was formulated that could quantify what catch sizes a fish population

could withstand without declining. Even though this concept of a maximum sustainable

yield (MSY) has been around for so long, fish stocks around the world have collapsed

due to overexploitation (Kuparinen and Merilä, 2007). I therefore did not just try and

summarize how the fishing industry could benefit from changing the management and

allow for fish to grow larger. It sounds like a strong incentive, let the fish grow large

and make more money with less effort, but so does the concept of MSY. Instead I will

try and review what responses we can expect from a size selective fishing regime in the

Baltic Sea. How will this influence the Baltic Sea, rather than how it may influence the

fishing industry itself. The use of wild fish stocks as a renewable resource may have

consequences on the targeted population that are hard to predict. Even if stocks are not

depleted, the removal of individuals may have far reaching effects. Population

responses depend, not only on the rate of exploitation, but also on the life history traits

of the targeted species, such as, age at maturity, recruitment, and growth rates. Fisheries

seldom target fish in all sizes so the induced mortality is not random amongst cohorts.

This could potentially have more long term effects on the exploited species. During the

last twenty years, the issue of genetic changes as a consequence of fishery induced

mortality has been given increasing attention and is believed to be connected to stock

collapses and slow growth rate of previously reduced populations (Andersen et al.,

2007). The dramatic decline in large cod (Gadus morhua) in the Baltic Sea has led the

Swedish foundation BalticSea2020 to launch a project in 2013 where fishermen and

scientist aim to develop fishing practises that allow cod to reach larger sizes whilst not

compromising the prerequisites for the fishing industry (BalticSea2020, 2013).

- 4 -

The recent history of the Baltic Sea ecosystem

The Baltic Sea has undergone several major ecological changes in the last hundred

years. Intense hunting during the first half of the last century dramatically reduced the

populations of seals. Toxic pollutants combined with unintentional mortalities from

fishing kept seal populations low for the second half. At such low levels, seals were no

longer the dominating top predator. Cod filled the space and increased in the absence of

seals (Österblom et al., 2007).

Figure 1 Relative abundance and modelled biomasses, from (Österblom et al., 2007)

The nutrient loads from human activities on land increased at the same time, changing

the Baltic proper from oligotrophic to eutrophic and thereby giving cod stocks the

bottom up support for extensive population growth. Cod fisheries got one record season

after the other until the collapse of the fisheries in the late 1980s (Fig. 2).

- 5 -

Figure 2 Baltic eastern Cod (subdivisions 25 & 26), spawning stock biomass and

fishing mortality (F= Total catch divided with mean stock size) 1996-2011, from ICES,

Advice 2012

.

Figure 3 Map of the Baltic Sea with ICES sub divisions, ICES

- 6 -

The increase in primary production led to hypoxia as bacterial decomposition used up

oxygen in the heavier, deep layer waters. This is the same waters where cod

reproduction is bound to take place due to the demands for specific levels of salinity.

Cod recruitment is now more dependent on the highly variable inflow of seawater

through the Öresund and Danish strait (Nissling and Westin, 1997). Anoxic bottoms

also prevents phosphorous from being trapped in the sediments, and thereby help

upholding the eutrophic state of the sea (Schenau and De Lange, 2001). Overfishing and

low recruitment led to low cod stock sizes so the second top predator was swept away

and the two clupeids, Baltic herring (Clupea harengus membras) and sprat (Sprattus

sprattus) came to dominate the ecosystem. The release of predation pressure in

combination with eutrophication resulted in a fourfold increase in sprat. Through

trophic cascades, one species of copepods was eliminated by increased predation from

sprat (Baum and Worm, 2009, Österblom et al., 2007, Möllmann and Köster, 2002).

High fishing mortalities in combination with some years of low recruitment and

potentially also egg predation from the clupeids have kept the cod stocks from reaching

previous levels. Seal populations have been on the recovery and their predation on cod

is increasing. Of these three major shifts, only the eutrophication is believed to be a

new stable state that cannot be reversed by, in this case, decreasing nutrient loads to

previous levels (Österblom et al., 2007). The Baltic Sea ecosystem is simple compared

to other marine environments in the sense that it is species poor due to the brackish

water. On the other hand the management is complicated, by species and stocks being

separated in time and space (Parmanne et al., 1994) much because of the gradient in

salinity. The high densities of sprat and herring for example, are concentrated north of

the areas where cod is present (Eero et al., 2012) so predator-prey interactions are

- 7 -

entirely different north, south, or in, the overlapping zone (fig. 4).

Figure 4 Spatial distribution of sprat, herring and cod from acoustic surveys 2012, from

ICES 2013

Selective fishing Fishing can induce mortality rates on the different cohorts of a population in various

ways. If a stock is exploited but not on the brink of extinction, the fishery is probably

Figure 5 Selectivity curves for trawls ans gillnets, from (Holst et al., 2002)

- 8 -

under some sort of regulation. Otherwise, the high efficiency of a modern fishing fleet

would soon deplete the stock. The regulations may be direct, limiting what sizes of fish

that are allowed to be caught e.g. a minimum catch size. For example, 100kt of cod per

year with a minimum size of 35cm. Indirect regulations instead determine how and

when the fishing may be conducted. For example with minimum mesh sizes on gillnets

and trawls. The combinations of the regulatory tools in use determine what fish are

likely to be harvested. The probability of a trawler to catch fish of a certain size

typically produces a sigmoid curve with low catch rates on smaller fish and high on

larger specimens (Fig. 5). The threshold for when the trawlers start catching fish is set

by mesh size and speed of the vessel. Escape windows in trawlers increase the

steepness of the slope, sharpening the precision of the gear (Madsen et al., 2002). The

usage of escape windows with at least 120mm openings has been mandatory in

Denmark since 2008 (ICES, 2010). Selectivity also relies on compliance for the

regulations. Manipulation of the gear can be extensive if regulations are not carefully set

(Suuronen et al., 2007). Trawlers are increasing their catches as part of the total catch in

the Baltic Sea as shown by the data on landings from the Swedish Agency for Marine

and Water Management (HaV) (Fig. 6). The selectivity of trawlers also vary with the

size of the vessel and the type of trawl in use (Tschernij and Holst, 1999).

Gillnets on the other hand, typically produce a hyperbole curve that moves horizontally

depending on mesh size (Fig. 5). A gillnet is most likely to catch fish with a size that

corresponds with the mesh size of the net. Any deviation from the optimum size will

cause the retention rate to drop. Larger fish can result in a higher economic yield per

unit weight and can, therefore, be the preferred catch. This may often be the case if the

fisheries are regulated by quotas or by unit effort. The prises per kg for a nine year old

cod is almost double that of a four year old cod for Norwegian fishermen (Diekert et al.,

2010). Spatial differences in the distribution of the cohorts can induce uneven mortality

- 9 -

rates even when selective gear is not used (Madsen et al., 2002, Sinclair et al., 2002). In

addition to selection for size, fishing may cause uneven mortality rates between species,

stocks, sexes, and in time and space (Zhou et al., 2010).

-

Figure 6 Landings by method for Baltic cod, eastern stock 1999-2011, Sweden. Data

from HaV.

Fishery induced mortality An increased density independent mortality on older or larger individuals in a

population can change what life history strategy that yields the highest number of

offspring, or fitness (Reznick et al., 2001). The reproductive success of fish usually

corresponds with size. Larger females can produce much more eggs (Kuparinen and

Merilä, 2007). But to grow large takes time and resources, recourses that instead can be

spent on reproduction. A high mortality rate on larger individuals acts as a selective

pressure that favours early reproduction and low size at maturity. When individuals are

unlikely to survive to reproduce at several occasions, selection favours the ones that

allocate resources in the production of gametes instead of somatic growth. The strategy

- 10 -

of early reproduction is favourable when predation is concentrated on larger specimens

(Wootton, 1998). The idea of genetic changes due to commercial fishing began with the

study of effects from natural predation. In a long term study on guppies, (Reznick et al.,

2001) switched the predators in two systems. One had a preference for large adults and

the other for small juveniles. The inverted mortality patterns resulted in a significant

life-history evolution. It was later shown that commercial exploitation of fish stocks

induced the same mortality patterns as natural predators and that the evolutionary

effects did not differ (Reznick and Ghalambor, 2005). For mortality to act as a selective

pressure that changes the allele frequencies over generations, i.e. evolution, some of the

variation in these traits needs to be genetic. The existence of genetic variations in size-

at-age has been found, for examples in populations of Atlantic cod (Olsen et al., 2004).

The reaction norm does also involve phenotypic plasticity that makes it more difficult to

distinguish the evolutionary changes. The secondary effects would be the same either

way but as long as the changes only include phenotypic plasticity, recovery is more

likely to be quick. The added mortality rates also need to be large enough if they are to

result in genetic shifts. In stocks that are reduced or held at low levels by fishing this is

likely to be the case (Andersen et al., 2007).

Figure 7 Mean weight index for Baltic cod of the western (WB) and eastern (EB) stock

1976-2010, from ICES (2013).

- 11 -

Using back calculated length-at-age data, from a close to 30 year period on cod in St.

Lawrence bay Canada, Sinclair et al. (2002) found that the fishing had led to changing

selective pressures over the period. A dome shaped mortality curve during high

intensity fishing, led to disruptive selection favouring both small and large size-at-age.

An overall selection for small size-at-age during a period of high fishing pressure is

argued to be the reason for the persistently slow growth recorded even when the fishery

has been brought to a complete halt (Sinclair et al., 2002). Recently, this fishing

induced evolution has caught the interest of behavioural ecologists. Passive gears like

gillnets or traps, depend on the fish to more or less catch themselves. This leads to that

there may be induced selection on behavioural traits as well. These traits can be

correlated with the ones discussed earlier since behaviour patterns like boldness and

high activity that are usually found in individuals with high growth rates, will also

increase catchability in passive gears (Uusi-Heikkilä et al., 2008).

The rate of genetic change

The Baltic cod (Gadus morhua) populations experience a high fishing mortality (Fig.

2). Along with variations in recruitment, commercial fishing is considered to be one of

the reasons for the collapse of the stock in the late 1980s. A reduction of size-at-

maturity has been recorded with a decrease of the length at which 50% of the females

are mature from 49.6 cm in the 1980s to 33.2 cm in 1996 (Cardinale and Modin, 1999).

Andersen et al. (2007) set out to find the strength of the evolutionary pressure on these

traits from the mortality caused by cod fishing in the Baltic Sea. They used the size at

maturity to calculate the fitness, using both the mortality from fishing and predation.

The heritability and variation in the population along with the expected lifetime

reproductive output of a female at present size compared with the optimum to get the

selection response. The evolutionary change in size-at-maturity was calculated to -36g

- 12 -

per generation under the present fishing pressure F=1 year -1

with a decreasing effect

down to zero at F=0.14 year -1

. The optimum size at maturation was estimated to be

250g less than observed. Three different management regimes where also tested for the

possible use to reverse the effects and thereby enlarging the yield. Making a reserve for

large fish or only target mature fish could not sustainably reverse the selective pressure.

Only when the overall fishing pressure was heavily reduced (F=0.1 year -1

) the model

predicted an increase in size at maturity for the Baltic cod (Andersen et al., 2007).

Other secondary effects

A various number of secondary effects can arise from the changes in life history traits

like size- or age-at-maturity. In addition to the intrinsic value of big fish, the loss of

them can be associated with a reduction in other services (Reznick and Ghalambor,

2005). Jorgensen et al. (2007) listed four categories of long term effects from reduced

size and loss of genotypic diversity respectively.

Reduced size Reduced diversity

Supporting services Supporting services

Loss of control of top down nutrient cycling,

population recovery potential.

Structure of ecological niches, metabolic

diversity

Provisioning services Provisioning services

Reduction of fisheries yield and stock

stability.

Reduction of fisheries yield and stock

stability.

Provisioning services Regulating services

Reduction of fisheries yield and stock

stability.

Resilience to environmental fluctuations

including climate change.

Regulating services Cultural services

Altered trophic interactions and control of pest

species.

Indigenous cultures, nature experiences.

- 13 -

Figure 8 Relationship between predator and prey size for cod using regression equations

from Scharf et al. (2000)

For cod to act as the predator on clupeids in the Baltic Sea, in a top-down predator

controlled ecosystem, the stocks of cod must consist of fish large enough to prey on the

herring and sprat. Cod begin to turn piscivorous when about 30cm of length (Pihl,

1994). The maximum prey size then increases linear with predator size much faster than

both minimum and mean prey size (Fig. 8). A decline in cod mean size, results in more

individuals of the same size provided unchanged stock estimates. The utilisation of

smaller prey can therefore increase, passing on the effects down the food chain (Beyer

and Lassen, 1994). Mean size of Atlantic herring has decreased over the last decades

(Cardinale and Arrhenius, 2000). International Council for the Exploration of the Sea

(ICES) argues that the decrease is a result of density dependent factors from the

increased sprat population that acts on herring while Beyer and Lassen (1994) on the

other hand discuss the effects of lowered predation (ICES, 2010). Cod can only prey on

- 14 -

the smaller, juvenile herring, so a reduced predation will include more small herring

individuals in the calculation of a mean size (Parmanne et al., 1994). Beyer and Lassen

(1994) did not include a reduction of cod in the effective size range for predatory

behaviour in their study.

Cod spawning in the Baltic Sea is restricted to the three deeper basins where salinity is

correct for egg buoyancy. But because the sea is stratified with a distinct halocline,

oxygen concentrations drop with depth, opposite of the salinity. That leaves quite a thin

layer where both salinity and oxygen are within range for egg survival. As a result, the

Bornholm basin is the only one that is still active (Neuenfeldt and Beyer, 2003)

Nissling and Westin (1997) found a positive relationship between egg buoyancy and

female age. Older females produced larger eggs with lower density which would make

them buoyant in less saline water where oxygen levels are normally higher. The

proportion of eggs produced by females being at least five years old was indeed

positively correlated to total recruitment. In addition to that, there is a positive

correlation between female body size and fecundity(Kuparinen and Merilä, 2007).

The need for repeated influxes of seawater contributes to the inherent variation in

population size for the Baltic cod (Scheffer et al., 2001). Österblom et al. (2007)

suggests that the reduction of long living predators in the Baltic e.g. cod and seals, has

made the system more prone to experience great fluctuations in primary productions.

The return of seals as the top predator in combination with clupeid predation on cod

eggs and larvae can potentially fixate the current situation of low cod stocks to a new

stable state. After that, a reduction in fishing mortality on cod would no longer be

enough to generate any dramatic change in cod populations (Scharf et al., 2000).

- 15 -

Gear selectivity for larger fish is an important management tool along with effort

control and quotas. Increasing the minimum mesh size in heavily exploited populations

will reduce the mortality on younger cohorts and increase the age at first capture while

at the same time, providing the means for higher economical yields by concentrating

fishing and handling time on the more valuable, large fish. Diekert et al. (2010)

simulated alternative harvesting policies for the North-East Arctic cod and found the

optimal harvest model to be far from the one currently in use by Norway and Russia. An

increase in mesh size for gillnets by 70mm to 236mm and 202mmfor trawlers would in

a few years more than double the net income from the fisheries. At the same time it

would help building up the stock from 2.4 to 6 million tons. In the optimal scenario,

effort would be made to catch at the age of 9 years but at the same time, the part of the

stock between age 11 and 15 would increase from 10 to 40%. Fishing mortality in that

simulation planed out on 0,17 after the initial years of building up the stock (Diekert et

al., 2010).

A study on the 50-year California Cooperative Oceanic Fisheries Investigations

(CalCOFI) larval fish record found strong evidence for that the variations in stock

abundances does not reflect variations in stock utilisation, instead it originates from

secondary effects caused by harvesting. Selective fishing leads to age-truncated

populations with altered intrinsic growth rates and age at maturation which can cause

instability (Anderson et al., 2008). Fluctuations in exploited stocks are undesirable for

the fisheries and can increase the risk of local extinction (Stenseth and Rouyer, 2008).

- 16 -

Reversing the evolutionary effects of selective fishing

In a recently conducted study on Atlantic silverside (Menidia menidia) the effects of

selective fishing was simulated for six generations. They selected for small, medium

and large fish respectively, resulting in twofold differences in mean weight and yield at

harvest. The harvest was then stopped for five generations during which a slow increase

in body size was observed for the populations that had previously been selected for

small size. Close to full recovery was estimated to take 12 generations. The populations

that had been under selection for large size did not show the same rate of recovery.

Applying this rate of recovery on wild populations that usually have longer generation

time, recovery can take decades (Conover et al., 2009).

An alternative to selective fishing is proposed by Zhou et al. (2010) who is one of the

few that are recommending less selective fishing gear and not the other way around.

The disadvantages born from truncating the populations are so excessive that he argues

that an ecosystem based fisheries management should aim to impose even fishing

pressures among species, stocks, cohorts, and sexes. As long as the catches are usable

for human consumption and by catches of sensitive species are controlled, the absence

of restrictions will compensate for the smaller revenues the fisheries get from not

catching the best paid specimens. Directing fishing to isolated spawning grounds will

make sure that only mature fish are targeted. This favours larger size-at-maturation

since large individuals will have higher fecundity (Kuparinen and Merilä, 2007).

- 17 -

Discussion

The variables needed for fishery induced evolution for smaller age-at-maturation and

size-at-age, all seem to exist for the cod fishery in the Baltic Sea. A Combination of

high exploitation rate and the use of highly size selective fishing gear (Fig. 2,5,6).

Trawlers are even catching an increasing part of the total catch (Fig. 6). And since the

retention rate for very large specimens in trawlers is higher than in gillnets (Fig. 5) the

situation does not seem to resolve by itself. This is supported by the mean size trend for

older cod (Fig. 7).

The secondary effects from smaller size-at-age, together with an age truncated

population, do go beyond the reduction of economical yields from the fishery. It

includes a decrease in the inherent stability of the population by reducing cannibalism

and by removing the part of the population most valuable for a successful spawning

(Vallin and Nissling, 2000, Neuenfeldt and Köster, 2000). Through trophic cascades,

the entire Baltic sea might be influenced from the reduction in large sized cod

(Österblom et al., 2007, Möllmann and Köster, 2002). In order to restore the ecosystem

functions of the cod as a top predator in the Baltic Sea, an increase in abundance may

not be enough. Larger mean size is not necessarily a sufficient criterion for solution

either. Such a measure would show positive results from a decrease in smaller

individuals (e.g. a decline in recruitment) as shown with the example of the declining

mean size of herring. An increase in predation from cod on sprat and herring can only

be achieved by an increase of large cod in the areas with high prey fish density (Eero et

al., 2012). Today’s minimum catch size of 38 cm results in a short time period for some

of the ecosystem services from cod, considering they begin to prey on other fish only

when they reach a length of about 30cm.

- 18 -

A substantial increase in the mesh size for trawlers would provide a temporarily relief in

fishing pressure and allow for an increase in population density. A total allowable catch

similar to the present would then inflict a lower fishing mortality and there by lessen the

evolutionary selection pressure for early maturation. Age at first capture would also

increase and larger individuals of cod would increase along with their predation on

clupeids and smaller cod. Cannibalism would select for later maturation and to a small

degree cancel out the selection pressure caused by fishing. It is important to remember

that the Baltic Sea is far from homogenous and that predation at the moment seems to

be limited partly by the geographic mismatch in predator-prey distribution. The

maternal effects on egg buoyancy and thereby on recruitment, is of great importance

given the unpredictable influxes of seawater. A long period without such an inflow is

likely to result in a drop in recruitment but an abundance of large females could temper

the effects (Nissling and Westin, 1997). Even after a large increase in mesh size, the

fishery would still inflict a selective pressure on the stock but with reduced power

(Andersen et al., 2007). The problem of lowering the fishing mortality can also be

solved by reducing effort or by creating no-take-reserves where no fishing takes place

but according to Conover et al. (2009), these reserves has to be very large to be

effective. Marine reserves do however, allow a high fishing pressure outside the

protected areas (Kuparinen and Merilä, 2007).

- 19 -

References

ANDERSEN, K. H., FARNSWORTH, K. D., THYGESEN, U. H. & BEYER, J. E.

2007. The evolutionary pressure from fishing on size at maturation of Baltic

cod. Ecological Modelling, 204, 246-252.

ANDERSON, C. N. K., HSIEH, C.-H., SANDIN, S. A., HEWITT, R., HOLLOWED,

A., BEDDINGTON, J., MAY, R. M. & SUGIHARA, G. 2008. Why fishing

magnifies fluctuations in fish abundance. Nature, 452, 835-839.

BALTICSEA2020. 2013. Towards length-optimised fishing in the Baltic Sea [Online].

BAUM, J. K. & WORM, B. 2009. Cascading top-down effects of changing oceanic

predator abundances. Journal of Animal Ecology, 78, 699-714.

BEYER, J. E. & LASSEN, H. 1994. The effect of size-selective mortality on the size-

at-age of Baltic herring. Dana, 10, 203-234.

CARDINALE, M. & ARRHENIUS, F. 2000. Decreasing weight-at-age of Atlantic

herring (Clupea harengus) from the Baltic Sea between 1986 and 1996: a

statistical analysis. ICES Journal of Marine Science: Journal du Conseil, 57,

882-893.

CARDINALE, M. & MODIN, J. 1999. Changes in size-at-maturity of Baltic cod

(Gadus morhua) during a period of large variations in stock size and

environmental conditions. Fisheries Research, 41, 285-295.

CLEGHORN, J. 1855. On the Causes of the Flutuations in the Herring Fishery. Journal

of the Statistical Society of London, 18, 240-242.

CONOVER, D. O., MUNCH, S. B. & ARNOTT, S. A. 2009. Reversal of evolutionary

downsizing caused by selective harvest of large fish. Proceedings of the Royal

Society B-Biological Sciences, 276, 2015-2020.

DIEKERT, F., HJERMANN, D., NÆVDAL, E. & STENSETH, N. 2010. Spare the

Young Fish: Optimal Harvesting Policies for North-East Arctic Cod.

Environmental and Resource Economics, 47, 455-475.

EERO, M., VINTHER, M., HASLOB, H., HUWER, B., CASINI, M., STORR-

PAULSEN, M. & KÖSTER, F. W. 2012. Spatial management of marine

resources can enhance the recovery of predators and avoid local depletion of

forage fish. Conservation Letters, 5, 486-492.

HOLST, R., WILEMAN, D. & MADSEN, N. 2002. The effect of twine thickness on

the size selectivity and fishing power of Baltic cod gill nets. Fisheries Research,

56, 303-312.

ICES 2010. Report of the Baltic Fisheries Assessment Working Group (WGBFAS), 15-

22 April 2010, ICES Headquarters, Copenhagen. . 633pp.

ICES 2013. Multispecies considerations for the central Baltic stocks: cod in

Subdivisions 25–32, herring in Subdivisions 25–29 and 32, and sprat in

Subdivisions 22–32.

JORGENSEN, C., ENBERG, K., DUNLOP, E. S., ARLINGHAUS, R., BOUKAL, D.

S., BRANDER, K., ERNANDE, B., GARDMARK, A., JOHNSTON, F.,

MATSUMURA, S., PARDOE, H., RAAB, K., SILVA, A., VAINIKKA, A.,

DIECKMANN, U., HEINO, M. & RIJNSDORP, A. D. 2007. Ecology:

Managing Evolving Fish Stocks. Science, 318, 1247-1248.

KUPARINEN, A. & MERILÄ, J. 2007. Detecting and managing fisheries-induced

evolution. Trends in Ecology & Evolution, 22, 652-659.

MADSEN, N., HOLST, R. & FOLDAGER, L. 2002. Escape windows to improve the

size selectivity in the Baltic cod trawl fishery. Fisheries Research, 57, 223-235.

- 20 -

MÖLLMANN, C. & KÖSTER, F. W. 2002. Population dynamics of calanoid copepods

and the implications of their predation by clupeid fish in the Central Baltic Sea.

Journal of Plankton Research, 24, 959-978.

NEUENFELDT, S. & BEYER, J. 2003. Oxygen and salinity characteristics of

predator–prey distributional overlaps shown by predatory Baltic cod during

spawning. Journal of fish biology, 62, 168-183.

NEUENFELDT, S. & KÖSTER, F. W. 2000. Trophodynamic control on recruitment

success in Baltic cod: the influence of cannibalism. ICES Journal of Marine

Science: Journal du Conseil, 57, 300-309.

NISSLING, A. & WESTIN, L. 1997. Salinity requirements for successful spawning of

Baltic and Belt Sea cod and the potential for cod stock interactions in the Baltic

Sea. Marine Ecology Progress Series, 152, 261-271.

OLSEN, E. M., HEINO, M., LILLY, G. R., MORGAN, M. J., BRATTEY, J.,

ERNANDE, B. & DIECKMANN, U. 2004. Maturation trends indicative of

rapid evolution preceded the collapse of northern cod. Nature, 428, 932-935.

PARMANNE, R., RECHLIN, O. & SJÖSTRAND, B. 1994. Status and future of

herring and sprat stocks in the Baltic Sea. Dana, 10, 29-59.

PIHL, L. 1994. Changes in the diet of demersal fish due to eutrophication-induced

hypoxia in the Kattegat, Sweden. Canadian Journal of Fisheries and Aquatic

Sciences, 51, 321-336.

REZNICK, D., BUTLER IV, M. J. & RODD, H. 2001. Life‐history evolution in

guppies. VII. The comparative ecology of high‐and low‐predation environments.

The American Naturalist, 157, 126-140.

REZNICK, D. N. & GHALAMBOR, C. K. 2005. Can commercial fishing cause

evolution? Answers from guppies (Poecilia reticulata). Canadian Journal of

Fisheries and Aquatic Sciences, 62, 791-801.

SCHARF, F. S., JUANES, F. & ROUNTREE, R. A. 2000. Predator size-prey size

relationships of marine fish predators: interspecific variation and effects of

ontogeny and body size on trophic-niche breadth. Marine Ecology Progress

Series, 208, 229-248.

SCHEFFER, M., CARPENTER, S., FOLEY, J. A., FOLKE, C. & WALKER, B. 2001.

Catastrophic shifts in ecosystems. Nature, 413, 591-596.

SCHENAU, S. J. & DE LANGE, G. J. 2001. Phosphorus regeneration vs. burial in

sediments of the Arabian Sea. Marine Chemistry, 75, 201-217.

SINCLAIR, A. F., SWAIN, D. P. & HANSON, J. M. 2002. Measuring changes in the

direction and magnitude of size-selective mortality in a commercial fish

population. Canadian Journal of Fisheries and Aquatic Sciences, 59, 361-371.

STENSETH, N. C. & ROUYER, T. 2008. Ecology: Destabilized fish stocks. Nature,

452, 825-826.

SUURONEN, P., TSCHERNIJ, V., JOUNELA, P., VALENTINSSON, D. &

LARSSON, P. 2007. Factors affecting rule compliance with mesh size

regulations in the Baltic cod trawl fishery. ICES Journal of Marine Science:

Journal du Conseil, 64, 1603-1606.

TSCHERNIJ, V. & HOLST, R. 1999. Evidence of factors at vessel-level affecting

codend selectivity in Baltic cod demersal trawl fishery. ICES CM, 2, 11.

UUSI-HEIKKILÄ, S., WOLTER, C., KLEFOTH, T. & ARLINGHAUS, R. 2008. A

behavioral perspective on fishing-induced evolution. Trends in Ecology &

Evolution, 23, 419-421.

VALLIN, L. & NISSLING, A. 2000. Maternal effects on egg size and egg buoyancy of

Baltic cod, Gadus morhua: Implications for stock structure effects on

recruitment. Fisheries Research, 49, 21-37.

- 21 -

WOOTTON, R. J. 1998. Ecology of teleost fishes, Kluwer Academic Publishers.

ZHOU, S., SMITH, A. D. M., PUNT, A. E., RICHARDSON, A. J., GIBBS, M.,

FULTON, E. A., PASCOE, S., BULMAN, C., BAYLISS, P. & SAINSBURY,

K. 2010. Ecosystem-based fisheries management requires a change to the

selective fishing philosophy. Proceedings of the National Academy of Sciences,

107, 9485-9489.

ÖSTERBLOM, H., HANSSON, S., LARSSON, U., HJERNE, O., WULFF, F.,

ELMGREN, R. & FOLKE, C. 2007. Human-induced trophic cascades and

ecological regime shifts in the Baltic Sea. Ecosystems, 10, 877-889.