Previous PageTable Of ContentsNext Page

Fisheries monitoring: Management models, compliance and technical solutions

Meryl J. Williams and Violeta P. Corral

International Centre for Living Aquatic Resources Management (ICLARM), McPO Box 2631, Makai City 0718, Phillipines.
email: [email protected]

Abstract: Monitoring must be considered within the broader context of fisheries management, which in turn is a part of a mosaic of multiple ocean and natural resource uses, including environmental conservation. Monitoring has a role to play in all aspects of fisheries management, including those related to the sustainable management of the resource, the economic performance of the fishery and the distribution of benefits from the exploitation of the resource and use of the environment. An immense challenge faced by management is that of finding cost-effective monitoring methods. Given its multiple roles, resources must be allocated to monitoring against competing but often related needs from other management related activities such as licensing, planning and legislation, policy formulation, marketing, surveillance and enforcement and research. This paper examines the challenges of fisheries monitoring and the extent to which solutions lie in (i) finding appropriate fisheries management models and plans, (ii) understanding the fishers motivation for compliance, and (iii) technical solutions.


  1. Monitoring involves the collection, measurement and analysis of fishing activity, including: catches, species composition, fishing effort, discards, area of operation, etc.
  2. Control involves the specification of terms and conditions under which resources can be harvested, which are normally contained in fisheries legislation and other arrangements (national, sub-regional, regional).
  3. Surveillance involves the checking and supervision of fishing activity to ensure that national legislation and terms, conditions of access, and management measures are observed.

Figure 1: Fisheries management and monitoring within the wider context of oceans management

Figure 2. Some dimensions of the differences between small and large scale fisheries, at a global scale (extracted from Thompson 1980).


Table 1. Hypotheses for the failure of fisheries management. The hypotheses marked in bold are those in which monitoring plays a significant role.




Data uncertainty

Simple models

Lack of ownership


Institutional frailty


Tuchman (1984): "... perverse persistence in a policy demonstrably unworkable."

Graham (1956): "inherent limitations of ... fishery statistics..."

Walford (1961): Fishing "cannot be understood out of context from the intricate system of their biological environment."

Beverton and Holt (1957): Perfection of regulation will require "some modification of .. individualistic and competitive approach."

Wilson et al. (1994): The "complexity and perhaps chaotic nature of the biological environment" makes management intractable.

Holt and Talbot (1978): "Institutions are imperfect."

Ludwig et al. (1993): "Short-sightedness and greed of humans underlie difficulties in management."

Source: Smith (1998)


3.1 Compliance behavior and legitimacy

3.2 Community based approaches.

3.3 'No force' approach

3.4 Advocacy approach

3.5 Oceans policy approach




McMANUS, J.W., ABLAN, M.C.A., VERGARA, S.G., VALLEJO, B.M., MENEZ, L.A.B., REYES, K.P.K., GOROSPE, M.L.G. & HALMARICK, L. 1997. Reefbase aquanaut survey manual. ICLARM Educ. Ser. 18, 61 p.

ABDULLAH, N.M.R., KUPERAN, K. V. & POMEROY, R.S. 1998. Transaction costs and fisheries co-management. Mar. Res. Econ., 13: 103-114.

ALVERSON, D., FREEBERG, M., MURAWSKI, S. & POPE, J. 1994. A global assessment of fisheries bycatch and discards. FAO Technical Paper No. 339. Rome, Italy. 233 pp.

BROWN, C. 1994. Overview of the Forum Fisheries Agency regional monitoring, control and surveillance program. In: ANON. Design and Implementation of a MCS System for the Philippines (Workshop Report), Manila, Philippines, 21-23 February 1994.

BRS 1998. Fishery Status Reports, Bureau of Resource Sciences, Canberra, Australia.

ENVIRONMENT AUSTRALIA 1998. Australia's ocean policy. Environment Australia, Canberra, Australia.

CULLET, P. & KAMERIE-MBOTE, A. 1996. Dolphin bycatches in tuna fisheries: A smokescreen hiding the real issues?. Ocean Dev. Int. Law (Ocean Development and International Law), 27:333-348.

DE LA MARE, W. 1998. Tidier fisheries management requires a new MOP (management-oriented paradigm). Rev. Fish Biol. Fish., 8: 349-356.

ENGLISH, S., WILKINSON, C. & BAKER, V. (eds.) 1994. Survey manual for tropical marine resources. Australian Institute of Marine Science, Townsville, Queensland, Australia.

FAO 1995. Code of Conduct for Responsible Fisheries. Rome, FAO. 41 p.

FAO 1998. State of world fisheries and aquaculture. Rome, FAO. 128 p.

DFO 1997. The role of the federal government in the oceans sector. Department of Fisheries and Oceans, Ottawa, Ontario, Canada.

HERSUOG, B. & PAULSON, O. 1996. Monitoring, control and surveillance in fisheries management, University of Namibia, Windhoek, Namibia.

ICLARM 1998 Reefbase. International Center for Living Aquatic Resources Management (ICLARM), Manila, Philippines.

KUPERAN, K.V. 1994. Enforcement and compliance with regulations in the Malaysian fishery. AFSSRN Research Report Series No. 3-3. AFSSR, IDRC and ICLARM.

KUPERAN, K .V. & SUTINEN, J.G. 1998. Blue water crime: Deterrence, legitimacy, and compliance in fisheries. Law Soc. Rev. (Law and Sociey Review), 32 :309-337 .

LEPIZ, L.G. & SUTINEN, J.G. 1985. Surveillance and enforcement operations in the Costa Rican tuna fishery. Mar. Pol.(Marine Policy), 9 :310-321.

MIDDENDORP, H.A.J., HASAN, R. & APU N.A 1996. Community fisheries management of freshwater lakes in Bangladesh. Naga ICLARM Q., 19 (2): 4-8.

MOORE, G. 1993. Enforcement without force: New techniques in compliance control for foreign fishing operations based on regional cooperation. Ocean Dev. Int. Law, 24:197-204.

NMFS 1997. Our Living Oceans. National Marine Fisheries Service, National Oceanographic and Atmospheric Administration, Office of Science and Technology, Silver Spring, Maryland, USA.

OCEAN VOICE INTERNATIONAL 1998. Trawling affects an area equal to half the world's continental shelves (News release), 15 December 1998. Ocean Voice International, Box 37026 3332 McCarthy Road, Ottawa, Ontario K1V 0W0, Canada.

PARSONS, L.S. 1993. Management of marine fisheries in Canada. National Research Council of Canada, Ottawa, Canada. 763 pp.

PITTENGER, R. F. 1997. Access to the sea: Encompasses ships, submersibles, autonomous and remote vehicles, observatories, drifters, extreme climate capability, and drilling. Oceanus Spring/Summer 1997.

POMEROY, R. S. & WILLIAMS, M.J. 1994. Fisheries co-management and small-scale fisheries: a policy brief. International Center for Living Aquatic Resources Management (ICLARM), Manila, Philippines. 15 pp.

ROGERS, D.R., ROGERS, B.D., DE SILVA, J.A. and WRIGHT, V.L. 1997. Effectiveness of four industry-developed bycatch reduction devices in Louisiana's inshore waters. Fish. Bull., 36: 552-565.

SCHUMAN, R. 1997. The future of regional fisheries cooperation in a changing economic environment: the South Pacific Island Countries in the 1990s. Ocean Dev. Int. Law, 28:369-403.

SECOND LONDON OCEANS CONFERENCE 1998. Report of a Workshop sponsored by the Governments of Brazil and the United Kingdom held at the Queen Elizabeth II Centre, London, 10-12 December 1998.

SILVESTRE, G. & PAULY, D. (eds.) 1997. Status and management of tropical coastal fisheries in Asia. Asian Development Bank and ICLARM, Manila, Philippines.

SMITH, T. D. 1998. Simultaneous and complementary advances': mid-century expectations of the interaction of fisheries science and management. Rev. Fish Biol. Fish., 8: 335-346.

STUMP, K. & D. BATKER 1996. Sinking fast: How factory trawlers are destroying US fisheries and marine ecosystems. Greenpeace Report. Washington D.C., USA.

THOMPSON, D. 1980. Conflict within the fishing industry, ICLARM News letter 3/3, ICLARM, Manila, Philippines.

VAN MULEKOM, L. & TRIA, E. C. 1997. Community-based coastal resource management in Orion (Bataan, Philippines): Building community property rights in a fishing community. Naga, ICLARM Q. 20 (2): 51-55.


Stock assessment of target species

Jake Rice

Co-ordinator, Canadian Stock Assessment Secretariat, Department of Fisheries and Oceans, 200 Kent Street, Stn 13W107, Ottawa, Ontario, K1A 0E6, Canada.
: [email protected]

Abstract: The management of fish stocks and fisheries depends crucially on the availability of reliable stock assessments. More specifically, a useful assessment usually must include reliable estimates of the quantity of fish in the stock, its age and/or size composition, its rate of renewal (recruitment and growth), and its rate of mortality (by fishing and by other causes). Typically these estimates are derived from two sources; research surveys and data from commercial fishing operations, combined with biological knowledge from targeted research programmes. In recent years, however, the reliability of assessments has been criticised in many regions, and unfortunately at least some of the criticisms have some validity. Even more regrettably, the states of fish stocks in many regions are poor or declining, and in many of those cases unreliable assessments are implicated as part of the problem. The serious consequences of what has become known as "the retrospective problem" within the stock assessment community is demonstrated to be widespread and prevalent even in countries which invest heavily in research and stock assessments. The collection of reliable data from fisheries operations is not just a good way, but possibly the only feasible way to correct the anomalies in the data associated with unreliable assessments. Accordingly, the features of data sets, and the biological and fisheries events which may cause a strong retrospective pattern to occur in data sets are examined and improvements to traditional assessment estimates illustrated through the use of high quality commercial data on catches, and on the technology and procedures used in harvesting. To view the value of data from monitoring programmes as simply an opportunity to improve traditional assessments, however, is to undervalue those programmes. Several illustrations are presented of how advice on management and conservation of fish stocks can be improved if assessments expand beyond their traditional catch-at-age basis. Data on the spatial and temporal dynamics of fisheries operations, in particular, may be vital to detecting important trends in stocks and fisheries. Moreover, the singular importance of complete and accurate data from monitoring processes and their contribution to significant improvements in both research surveys and targeted scientific research programmes is illustrated. Finally, contemporary fisheries science and management has historically progressed beyond single-species assessments and management initiatives, to consider multi-species and ecosystem interactions, and the effects of the physical environment on marine populations. Although outside the subject matter of this discussion, sound monitoring programmes can contribute greatly to our knowledge in this larger framework, as well as contributing to improved single-species assessments of target species.


Management agencies have tried a vast array of tools, in their efforts to manage the exploitation of fish stocks. Where agencies have had the management authority for a fishery and sufficient capacity has existed that a fishery had the potential to become over-exploited, management generated restrictions have generally been implemented (e.g. sizes of catches, sizes of fish, total effort, numbers of participants, catches per trip, gear characteristics, bycatches, and just about everything else which could be measured and limited (Mace 1996, Hey 1996). Despite all these efforts at control, fisheries and fish stocks internationally are not thriving (Alverson et al. 1994, Meltzer 1994, Alverson 1997, Mace 1997). There are some informative patterns to be observed in this history, however.

Typically, each new management tool comes with great publicity, great expectations, and great promises. However, after a suitable interval in which to evaluate the actual effectiveness of the tool, the initial optimism is replaced with more realistic claims: usually that the measure did some good, but offered no total solution. It has been argued that managers should be adaptive (Walters 1986, Hilborn and Walters 1992), but fishers repeatedly show that they, too can be highly adaptive (Rice and Richards 1996). This dynamic cycle of conservation concern, management measures, and industry adaptation is another common pattern through history.

Rice and Richards (1996) discuss the common dynamic interaction between regulatory regimes and activities of fisheries, highlighting the non-functionality of the dynamics when regulators and harvesters do not share common objectives. Here, we should step back and consider the question of why users and managers should ever fail to share core objectives to begin with? At the most fundamental level, fisheries scientists, management agencies, and resource users all must want viable fisheries in ecosystems which are being sustained. What differs between them is what each believes must be done to achieve those two goals. Hence, we arrive at another consistent feature of fisheries management systems (sensu Stephenson and Lane 1995): management measures work best when resource users and resources managers both have confidence in first the evaluation of the present state of the stock(s) and fisheries, and second the measure(s) necessary given that stock state.

Why do one or both groups fail to have such confidence in the evaluation of the stock(s)? There are many candidate reasons. Trade journals are filled with articles illustrating that resource users and managers can have different perceptions of the consequences of a given management measure (for examples from one year's issues of one quite responsible Canadian industry publication, see Mackinson 1998, Pepper 1998, Spisak 1998, Stephenson 1998). Single sector publications can be much more strident. Users feel they are being forced to live under regulatory regimes designed by bureaucrats lacking valid knowledge of the realities of the fisheries. As for confidence in the evaluations of the state of the resources, it is now a "received truth" that stock assessments are at best unreliable and often simply the product of faulty methods applied by incompetents to unreliable data (ex. Martin 1995). Hence, another historical generalisation: to correct any of the major ills in fisheries and ecosystem management, one has to start with the accurate measurement of parameters. Without accurate measurements one cannot even begin to sort out the correct methods and the competent assessors, which is a necessary precursor to the task of creating a single perception of stock status shared by all parties.

This paper describes the value of monitoring programmes to stock assessment. The value is obvious. Monitoring provides data, and data are the raw material for stock assessment. No monitoring - no assessment. Although this appears to be a terminal statement, there is much more to this link between monitoring and assessment. Exactly how have unreliable data contributed to the inaccuracy of past stock assessments, and as a consequence to the crises in the world's fisheries? How widespread is the problem of unreliable catch data? Is it universal, or only characteristic of regions of the world which cannot (or choose not to) invest heavily in fisheries science and management? What are the damaging properties of the historic data sets? This is important because if it were just that fishery data were noisy; then estimates would only be highly uncertain, with large confidence intervals. However, the historic data sets associated with the most problematic assessments may have characteristic pathologies, leading to particularly insidious inaccuracies in assessments using the flawed data. Can such pathologies be improved or even eliminated through effective monitoring programmes?

The value of monitoring programmes to stock assessment does not end with improved data, however. Once the data are reliable, it is possible to evaluate the assessment methods as well. However complex a mathematical assessment model is, it still remains just a series of statements about the biology of the fish, and the operations of the fishery, both converted into mathematical algorithms. The knowledge gained from effective monitoring can make these mathematical expressions better in many ways. This can improve the assessment methods, as well as the data which go into them.


Although researchers constantly plead for good data developments over the past decade have made it possible to quantify just how serious the problems with data quality are. This is of particular importance with regard to inaccuracies in assessments and scientific advice on management deriving from poor quality data on commercial catches.

Since the mid-1980s, the "retrospective problem" in assessments has been receiving much attention (summaries see Mohn 1993, 1999, Mesnil 1995). When one conducts an assessment in year x, using catch and research data up to at least year x-1 (and often, now, year x as well), typically several key estimates and forecasts are produced:

  1. The total and spawning biomasses on January 1 of year x+1 and all previous years;
  2. The exploitation rate (often expressed as fishing mortality) for the current year and all previous years, derived from ratios of the reported catches and estimated biomasses, and
  3. An estimate of the catch for year x+1 which would correspond with the management objective through the application of a target exploitation rate and the January 1 biomass in year x+1.

From a management perspective the assessment is undertaken primarily to obtain the forecasts for January 1, year x+1, but each assessment updates estimates for the full time series. It has been known for decades that sequential population analyses converge over time (Ricker 1975), so our perception of the history of a stock becomes quite stable. However, frustratingly often, when the process is repeated the next year (during year x+1, for forecasts of January 1, year x+2 and all previous years), those estimates of the biomass and exploitation rates in recent years differ in systematic ways from the estimates of the same entities obtained in year x. This is the "retrospective pattern". It means that the management during that year has been based on false premises, and management objectives have not been achieved. Often the pattern is repeated for several successive years.

How widespread and how serious is this retrospective problem? Consider the set of groundfish assessments conducted by the International Council for the Exploration of the Sea (ICES) in the autumn of 1998 for fish stocks in the North Sea, Barents Sea, and northern shelf of the European Atlantic coast. These assessments have been chosen because of the availability of the necessary documentation to the author, and because of the fact that these assessments, for the following reasons, should be among the best family of assessments conducted by the fisheries science community:

  1. Data is collected by competent national and international fisheries management agencies where there are serious efforts to manage the fisheries and to provide catch statistics;
  2. The data from these fisheries are augmented by abundant survey data of high quality;
  3. The biology of the species is known well;
  4. ICES draws on an extremely skilled and experienced community of assessment scientists for preparing and reviewing the assessments, and
  5. The time series of catches and survey data are relatively long.

To gain some consistent perspective on the course of retrospective changes over time, this paper considers the estimate of fishing mortality in year 1994, from assessments using the most current data set, but truncated in 1995, 1996, 1997, and 1998. Table 1 presents the estimates of F in 1994 over that period. For some stocks, such as saithe and sole in the North Sea, our perception of the exploitation intensity has changed very little. For other stocks, such as most of the whiting stocks, our estimates have changed greatly. It is of interest to note that although the larger changes do not appear to be random, the apparent bias is in the same direction for several successive years. The direction of this bias is not consistent, however, for all stocks. In some cases, such as North Sea plaice and Barents Sea haddock, assessments have become generally more optimistic rather than pessimistic about the state of a stock,

Table 1. Estimates of fishing mortality in 1995 for groundfish stocks in the Northeast Atlantic, if the most current catch and survey data series is truncated in 1995, 1996, and 1997. Data from ICES 1999 a, b, and c. It is not possible to estimate f analytically for Norway pout, so estimates are of Spawning Stock biomass.

Stock name (age range) Data series truncated in:
  1995 1996 1997
North Sea cod (2-4) 0.64 0.68 0.78
North Sea Haddock (2-6) 0.67 0.74 0.85
North Sea whiting (2-6) 0.47 0.58 0.73
North Sea saithe (3-8) 0.58 0.58 0.56
North Sea sole (2-8) 0.51 0.46 0.52
English Channel sole (3-8) 0.59 0.41 0.28
North Sea plaice (2-10) 0.43 0.45 0.39
Skagerrak plaice (4-8) 0.86 1.08 0.81
English Channel Plaice (2-6) 0.59 0.57 0.40
North Sea Norway pout (SSB in t ) 1.6 x 105 1.7 x 105 2.2 x 105
Barents Sea cod (5-10) 0.51 0.69 0.78
Barents Sea haddock (4-7) 0.41 0.31 0.30
Celtic Sea whiting 0.88 0.77 0.62
Whiting in subarea VIa 0.92 0.81 0.79

It is not the intention of this paper to single out the ICES community. Retrospective patterns are comparably common and serious in Canadian Atlantic groundfish assessments. Because many stocks are in a state of collapse, current analytical assessments are not always available and the illustrations have to come from different years. The small sample of data detailed below shows the magnitude of the problem, particularly during the late 1980s and early 1990s, when the stocks were on the verge of collapse (Table 2, Fig 1). The overly optimistic assessments are recognised as an important contributor to the eventual collapse of many of these cod stocks (Parsons 1993, Rice in press).

Table 2. Retrospective analysis of some Canadian Atlantic cod stocks. Data from CSAS and ASAS research documents and assessment authors (see www// Because of changes in analytical approaches to individual stocks over the 1990s, the target year, range of years, and estimated attribute [Attr: f = fishing mortality; n = 3+ numbers (,000); r = millions of recruits at age 2] varies among stocks.

Stock Attr. Target
Years of Estimates Value of the estimate in each year
2J3KL F 1989 1989-91 0.42 0.58 0.89 na
3Ps N 1993 1994-97 38,000 52,000 82,000 62,000
3Pn4RS N 1992 1993-96 21,000 28,000 30,000 30,000
4TVn F 1991 1992-95 0.34 0.46 0.53 0.54
4Vn(summer) F 1991 1992-95 0.28 0.19 0.35 0.41
4VsW F 1992 1993-96 1.72 1.14 1.23 1.30
4X R 1992 1994-97 17 23 28 25
5Zj,m R 1992 1994-97 1.5 1.9 2.9 4.2

It is important to note that the preceding tables under-represent the potential magnitude of the retrospective problem. These studies were conducted using exactly the same model formulations, and merely terminating the input data series at different years. Real assessments are done annually, and in successive assessments, there are likely to be methodological differences. The analytical formulations often change at least in detail from one meeting to the next, as do the relative mathematical and interpretational weightings given to different aspects of the data and the fits between population reconstructions and data sets. Consider Barents Sea cod, supporting one of the greatest cod fisheries in the world. The management target of Fmed (ICES in press), is generally considered sufficiently conservative, and simulations indicate this rate of exploitation is very likely to be sustainable (Gabriel 1994, Mace 1994, ICES 1997a, 1997b, 1998a). The successive assessments in 1996, 1997, and 1998 (ICES 1997c, 1998b, in press) have reduced the estimate of the January 1, 1997 SSB by over 40%, and the 1997 catch corresponding to the management target by over 50% (Table 3a). This is not an isolated case; changes to the estimates of Northern cod in 1987, from the 1989 to 1992 assessments were of comparable magnitude (Table 3b, CAFSAC 1990, 1991, 1992, 1993).

Table 3a. Estimates of Spawning Biomass (SSB x103 t) and target catch (x103 t) consistent with Fmed, for the year 1996, from assessments of Barents Sea cod in 1996 (ICES 1997c), 1997 (ICES 1998b), and 1998 (ICES in press).

Assessment Year SSB Target Catch
1996 1,300 994
1997 839 514
1998 762 478

Figure 1. Illustrations of retrospective patterns from current assessments of six Canadian Atlantic cod stocks for 3+ biomass (2J3KL, 3Ps, 3Pn4RS) or f [4TVn, 4Vn (summer), 4VsW]. Except for 3Pn4RS, the retrospective analysis is not of the final formulation accepted by the peer review body, either because additional components of the stock could not be concluded in the single analytical framework (2J3KL, 3Ps), or input patterns of natural mortality over time were changed in the final runs, and retrospective analyses were not repeated with the revised formulations.

Table 3b. Estimates of 3+ biomass, SSB (x103 t) and f from assessments of cod on NAFO Div. 2J3KL in 1988, from assessments in 1989, 1990, 1991, and 1992 (CAFSAC 1990-1993, respectively).

Assessment Year 3+ biomass SSB (7+) f
1989 1,050 453 0.44
1990 840 381 0.57
1991 820 387 0.58
1992 690 110

na (high)


For the past decade fisheries scientists have been searching for the causes of the retrospective problem, early warning signs that it may be present in an assessment, and methods to correct assessments for it when it is present. Although there is not yet full consensus on the details of any of these points, several generalisations are emerging (Mohn 1999). For our purposes, it is the potential causes which matter. It has been known since the 1980s that retrospective errors occur when an element or elements are systematically changing in the input data series, or there is a systematic change in the biology of stock being assessed (CAFSAC 1989). As noted earlier, the systematic aspect of such change is important as data which are just noisy simply lead to unbiased estimates with very high uncertainty.

In a fishery with a constant rate of discarding or mis-reporting the estimates of biomass and fishing mortality from an assessment based on the reported (but incorrect) catch will scale incorrectly relative to the absolute values of the stock, but trends are estimated reliably. If the management measures for the stock make the biomass trajectory stable or increasing, and the exploitation rate stable or decreasing, the stock is being sustained. This will be true even though the estimates of biomass and fishing mortality are incorrect.

For the terminal year the assessment process estimates the number and biomass of fish in each age class at the beginning of year, and decreases these numbers through the year, to account for the fish which will die from fishing activity and from natural causes. This produces the forecast population at the beginning of the succeeding year. If the rate of mis-reporting increases and the reported catch data from that year are put into the next assessment, the relative age composition shows fewer fish than were estimated for every cohort which was being discarded at the higher rate. The consequence is that the assessment estimates that the population at the start of the previous year was smaller than had been estimated last year, and, given the catch, the exploitation rate was higher than the target. This pattern continues for several years, until the assessment converges on a new relative scale. Throughout all that time management of the fishery is continually failing to meet its objectives. (If the rate of discarding or mis-reporting goes down, the effects are the reverse and the conclusion that the stock was under-estimated, and the exploitation rate over-estimated remains. This may be good for the stock, but it is often unpopular with resource users who must comply with management limits which appear to be more strict than necessary to achieve agreed management targets.)

Changes in mis-reporting or discarding rates are not the only possible cause of a retrospective pattern. Changes in biological properties such as natural mortality or growth can also contribute to the pattern. In practice, of course, it is extremely difficult to partition mortality between fishery induced and natural causes (Bax 1994, Mertz and Myers 1997). To the assessment a fish is equally dead from starvation, a predator, or unreported death in fishing gear. What the sequence of assessments demonstrates is that the age composition of the stock keeps becoming younger than estimated from the previous assessment, which in turn means there were fewer fish and they were surviving more poorly than had been estimated. (Conversely, in cases of decreasing mortality, there are more fish than thought, and they are surviving better.) It is up to the scientist to determine what has changed about the stock and why.

Changes to fishing practices can also contribute to retrospective and current model inaccuracies. Assessment models often use catch-per-unit-of-effort (CPUE) to calibrate relative trends and population values to some absolute scale, fixed by the absolute catches of commercial vessels. The problems with using CPUE are well known (Walters and Maguire 1996, Hilborn and Walters 1992), and in recent years, methodological alternatives have been developed (ICES 1993, 1995a, NRC 1998). These methods have their own problems, though (ICES 1995b, NRC 1998), and CPUE remains an attractive property to scale modelled populations to real ones. The use of CPUE, however, assumes that units of effort remain constant over time and does not account for changes to gears, vessel efficiency, or an increasing knowledge of the resource user on how to harvest their target stocks. All these changes alter the absolute catch per hour of fishing, per 1000 hooks, per metre of net, or by whatever catch unit is used. Such changes alter the scaling between the index and the population, resulting, for example, in an increase in efficiency falsely appearing as an increase in the stock (Walters and Maguire 1996). Again, this is not a unidirectional problem. In recent years industry has adopted many gear modifications to improve conservation aspects of harvesting. If such modifications lessen efficiency somewhat, this could change the scaling in the other direction, leading to a false suggestion that the stock is not benefiting sufficiently from the conservation measures, or that the measures are overly costly in terms of foregone catch.

Even if one is not using CPUE, changes to fishing practices can still contribute to inaccurate assessments. If a fishing gear or the prosecution of a fishery, in space or time, changes in ways which affect the age composition of the catch, or cause a different portion of the stock to be exploited, models assuming constant selectivity can go awry, sometimes badly (Casey 1996, Dealtaris and Riedel 1996, ICES 1998c). These effects again often appear as a retrospective pattern. The assessment process, in such circumstances, forecasts an age composition for the population and catch in the management year, assuming historic age selectivity for the fishery. When the catch data and the next assessment's population are reviewed, there are systematic differences in the age composition, which lead the assessors to conclude the past assessment was in error. This again continues until the several most recent years in the data series reflect catches from the new selectivity pattern.

In these two cases the absence of ancillary information about the fishery or stock underlay the inaccuracies in the assessments rather than the unreliability of the catch data. Mathematical models which reconstruct the population, or which estimate parameters of it, make specific assumptions about the biology of the species, and activities of the fishery. When natural mortality or selectivity change, or the fishery begins to exploit a different portion of the stock, these assumptions become invalid. These assumptions can be changed easily enough. The problem is to know when change has occurred, and in what way.


Monitoring fisheries will provide data on commercial catches, which are representative and reliable (as long as the monitoring programmes are designed and implemented competently). Simply by providing complete data on catches, bycatches, and discards consistently over a period of time, assessments will be able to begin with reliable data. Improvements in the quality of traditional commercial data will eliminate one source of inaccuracy in assessments; a source of inaccuracy which in simulations (where the "true" population trajectories are known, unlike the trajectories of actual fish stocks) often dominates all results, and is often a major contributor to large retrospective patterns.

More reliable data therefore, results in more accurate assessments. Whether the assessments have higher or lower variance (i.e. whether or not population estimates, and estimates of yields given target exploitation strategies, have broader or narrower confidence regions) depends on many details of the fishery and the monitoring programme. Even if the uncertainty of estimates is not changed, though, the potential for bias will be greatly reduced. As it is often the use of biased estimates which lead to particularly poor management decisions the importance of competently designed and implemented monitoring programmes is essential. As a result the traditional scientific basis provided to fisheries managers as an objective component of their decision-making will be more sound which should lead to better decisions by fisheries managers and improvements in both the sustainability of stocks and the economic viability of fisheries.

It may be a false hope that improved data immediately lead to more accurate analytical assessments, because the mathematical models themselves may be more poorly structured than envisaged. However, improved data are, by definition, a better reflection of the actual events in a fishery and the biological population being sampled. The scientific advice arising from an assessment is, in consequence, more than just the results of the mathematical computations and also contains an interpretational component. This interpretational component is very flexible, and it can change faster or slower than the actual annual computational results change (Finlayson 1994). It is possible that as the traditional data become more and more reliable reflections of events in exploited fisheries and populations, the data alone will guide the scientific community to more accurate interpretations of model outputs. In the medium term the improved interpretations of events will prompt changes to the model formulations as well, replacing erroneous formulations with ones more consistent with the improved interpretation of reality. So, even in this completely scientific loop, monitoring programmes necessarily lead to improved scientific advice, and more reliable mathematical formations of fisheries and biological processes.

As well as providing more reliable, traditional data integrated monitoring programmes provide new kinds of data as well and allows more complete recording of fisheries activities and vessel characteristics (e.g. gears used, exact places and times fished, and vessel features). One immediate benefit of such an approach is protection against undetected changes in fishing practices affecting gear selectivities, which is another common contributor to strong retrospective patterns. Other benefits include the ability to scale effort to something more than vessel length or horsepower, and look at catch rates more finely than perhaps days fished, providing additional fishery based indices of stock status. It may be less obvious, but also true, that integrated fisheries monitoring can allow a wider range of biological attributes of the catch to be measured. This gives us an opportunity for early detection of changes in growth and mortality rates, again increasing our protection against undetected retrospective assessment errors. Further consequences of improved monitoring practices include not just better measures of traditional parameters, but accurate measures of new parameters as well. These will allow even swifter and larger improvements to model formulations, replacing coarse surrogates of fishing capacity or power with more meaningful measures.

Another important consequence of improved advice will be the opening of the scientific loop. As the scientific advice becomes more accurate, the reasons for industry to distrust it may be reduced. Input data will be recognised as more reliable and through the processes explained above, the interpretations and model formulations are likely to become more credible as well. As the assessment data and models become more credible, we can hope that the distrust between sectors is another casualty of the monitoring programmes. When rational individuals can see that the model formulations of fishing activities are based on, and fit to, accurate representations of the true activities, they should distrust the results much less. Less distrust should lead to a higher likelihood of compliance with management plans based on the results. Better compliance will feed back to further improvements in data quality and comprehensiveness. Moreover, when the scientific and management communities have confidence in the data and models, the temptation to produce conservatively biased "precautionary" results or management implementations (see discussion in ICES 1997b, 1998a) is lessened, removing another source of distrust among parties.

Continuing the stepwise evaluation of benefits, the diminishing distrust among the scientific, management, and industry sectors produces its own benefits. As the marine ecosystem and the fishery-ecosystem interactions are complex (ICES 1995b, 1998d) it is unreasonable to expect integrated fisheries monitoring programmes to provide the basis for completely reliable analytical assessments in every case. Consequently alternative hypotheses about fishery and biological processes are a major component of many assessment models, and a major source of uncertainty in results.

The effects of better data and better model formulations may allow some alternative hypotheses to be rejected, and reduce the overall uncertainty about stock status and effectiveness of management actions. A more important contribution of integrated fisheries monitoring programmes is perhaps, the fact that, through greater trust and interaction amongst and between sectors, new and better hypotheses may be incorporated into fisheries models. Knowledgeable industry members can provide highly insightful hypotheses about marine ecosystems, and about ecosystem-fishery interactions. In this instance also monitoring programmes may lead to new processes being incorporated into assessment models as well as increasing the accuracy of traditional model formulations.


The implications of an effective integrated monitoring programme on the historical patterns observed in fisheries and assessments would be

  1. More reliable data for inputs to assessments;
  2. Development of better models for assessment computation. Models can be assessed and be made more complete and realistic;
  3. Increased trust can be developed between scientists, managers, and industry. The greater accuracy of assessment computations, and improved model formulations can make assessment results correspond more closely with industry's experiences, increasing their confidence in the assessments. At the same time the objectivity and empiricism of the monitoring programmes means that scientists and managers have greater confidence that the information they are working with is actually what the industry is doing. Moreover, the very nature of a monitoring programme provides a forum for the sectors to work closely together, and the common toil provides opportunities to break down unhelpful stereotypes that each sector may have of others, and
  4. Common objectives are required between parties. Superficially, monitoring programmes do not address objectives of scientists, managers, and fisheries. At a fundamental level, however, all parties must share broad core objectives of viable fisheries in sustained ecosystems. Problems arise when each party considers, on its own terms, whether a particular suite of measures of stock status and fisheries performance reflect viability and sustainability, and if not, what must be done to get there. It should be plausible that when all parties work with and trust common data, and models contain the processes which all parties agree are relevant and important, that there are many fewer disagreements about whether viability and sustainability are being achieved, and about the consequences of measures proposed to achieve or maintain those overall objectives.

New management approaches come with great fanfare, which fades as the promise is not delivered. Although integrated fisheries monitoring could be another such case, this is not the view of the author. Effective monitoring to obtain comprehensive and reliable data is not a new idea, and is not founded on novel theories of either ecology or fisheries. Rather, we are addressing the basics with objectivity, rigor, and common good will. That's a good foundation for a good future.


I would like to thank the many assessment scientists and biologists in Canada and the ICES community, whose work provided the material on which this paper is based, and who continue to do outstanding professional work with raw materials often of appalling quality. I would like to thank many individuals in the British Columbia Deep-Sea Trawlers Association, from whom I learned how much industry can help to improve scientific hypotheses as well as just provide data. I would particularly like to thank the ICES General Secretary and Fisheries Advisor for permission to draw heavily from analyses in Working Group Reports.


ALVERSON, DL. 1997. Global trends: Fisheries management. Am. Fish. Soc. Symp., 20: 343.

ALVERSON, D.L., FREEBERG, M.H, MURAWSKI, S.A. & POPE, J.G. 1994. A global assessment of fisheries bycatch and discards. FAO Fisheries Technical Paper 339, FAO, Rome, 233pp.

BAX, N.J. 1994. Estimation of natural mortality in ecological models. In : Population dynamics for fisheries management, p 63-70. Australian Society for Fish Biology, North Beach, Western Australia, Australia.

CAFSAC 1989. Report of the Statistics, Sampling and Surveys Subcommittee of CAFSAC, January, 1989, 339 pp.

CAFSAC 1990. Canadian Atlantic Fisheries Scientific Advisory Committee Annual Report 12, 1989, 306 pp.

CAFSAC 1991. Canadian Atlantic Fisheries Scientific Advisory Committee Annual Report 13, 1990, 252 pp.

CAFSAC 1992. Canadian Atlantic Fisheries Scientific Advisory Committee Annual Report 14, 1991, 204 pp.

CAFSAC 1993. Canadian Atlantic Fisheries Scientific Advisory Committee Annual Report 15, 1992, 217 pp.

CASEY, J. 1996. Estimating discards using selectivity data: The effects of including discard data in assessments of the demersal fisheries in the Irish Sea. J. northwest Atl. Fish. Sci., 19: 91-102.

DEALTERIS, J & RIEDEL, R. 1996. Effect of size selection within and between fishing gear types on the yield and spawning stock biomass per recruit and yield per unit effort for a cohort of an idealised groundfish. J. northwest Atl. Fish. Sci., 19: 73-82.

FINLAYSON, A.C. 1994. Fishing for truth: A sociological analysis of northern cod stock assessment from 1977 to 1990. Soc. Econ. Stud. No. 52; Inst. Soc. and Econ. Res. Mem. Univ. Nfld.

GABRIEL, W.L. 1994. A simple method for estimating uncertainty associated with Fmed commercial fish stocks. ICES C. M. 1994/S: 13: 13 p.

HEY, E. 1996. Global fisheries regulations in the first half of the 1990s. Int. J. Mar. Coast. Law, 11; 459-490.

HILBORN, R. & WALTERS, C.J. 1992. Quantitative fisheries stock assessment. Choice, dynamics and uncertainty. Chapman and Hall, New York, U.S.A., 570 pp.

ICES 1993. Report of the Working Group on Methods of Fish Stock Assessment. ICES CM 1993/Assess:12, 239pp.

ICES 1995a. Report of the Working Group on Methods of Fish Stock Assessment. ICES CM 1995/Assess:11, 188pp.

ICES 1995b. Report of the Working Group on Ecosystem effects of Fishing. ICES CM/ACFM/ACME: 1995/1. 212 pp.

ICES 1997a. Report of the Comprehensive Fishery Evaluation Working Group. ICES CM 1997/Assess, 15: 140 pp.

ICES 1997b. Report of the Study Group on the Precautionary Approach to Fisheries Management. ICES CM 1997/Assess, 7: 42 pp.

ICES. 1997c. Report of the ICES Advisory Committee on Fisheries Management, 1996. ICES Coop. Research Report 221, 2 Vols: 314 pp.& 327 pp.

ICES 1998a. Report of the Study Group on the Precautionary Approach to Fisheries Management. ICES CM 1998/Assess,10: 40 pp.

ICES 1998b. Report of the ICES Advisory Committee on Fisheries Management, 1997. ICES Coop. Res. Rept 223, 2 Vols: 413 pp.& 356 pp.

ICES 1998c. Report of the Study Group on the Use of Selectivity and Effort measures in Stock Assessment. ICES 1998/CMB: 6; 77 pp.

ICES 1998d. Report of the Working Group on Ecosystem effects of Fishing ICES CM/ACFM/ACME:1998/1. 387 pp.

ICES 1999a. Report of the Working Group on the Assessment of Demersal Stocks in the North Sea and Skagerrak. ICES CM 1999/ACFM:8; 687 pp.

ICES 1999b. Report of the Arctic Fisheries Working Group. ICES CM 1999/ACFM:3; 276 pp.

ICES 1999c. Report of the Working Group on the Assessment of Southern Shelf Demersal Stocks. ICES CM 1999/ACFM:4; 692 pp.

ICES. (in press). Report of the ICES Advisory Committee on Fisheries Management, 1998. (1998) ICES Coop. Res. Rept. 225.

MACE, P.M. 1994. Relationships between common biological reference points used as thresholds and targets of fisheries management strategies. Can. J. Fish. Aquat. Sci., 51: 110-122.

MACE, P.M. 1996. Limited access for Atlantic highly migratory fish stocks: Changing the ground rules. Fisheries, 21 (4); 20-22.

MACE, P.M. 1997. Developing and sustaining world fisheries resources: The state of the science and management. CSIRO, Collingwood (Australia); 1-20 pp.

MACKINSON, S. 1998. Diversity and Context in Managing Herring stocks: Observations and Opinion of the Central Coast Seine Fishery. West Coast Fisherman, 13: 22-24.

MARTIN, C. 1995. The collapse of the Northern Cod Stocks: Whatever happened to 86/25. Fisheries, 20: 6-9.

MELTZER, E. 1994. Global overview of straddling and highly migratory fish stocks: The non-sustainable nature of high seas fisheries. Ocean dev. Int. Law, 25: 255-344.

MERTZ, G. & MYERS, R.A. 1997. Influence of errors in natural mortality estimates in cohort analysis. Can. J. Fish. Aquat. Sci., 54: 1608-1612.

MESNIL, B. 1995. Retrospective analysis of stock assessment results. Les recherches Francaises en évaluation quantitative et modelisation des ressources et des systèmes halieutiques. Orstom, Paris (France). [Colloq. semin. inst. Fr. rech. Sci. dev. coop. orstom] 85-92 pp.

MOHN, R.K. 1993. Bootstrap estimates of ADAPT parameters, their projection in risk analysis and their retrospective patterns. Risk evaluation and biological reference points for fisheries management. Can. Spec. Publ. Fish. Aquat. Sci., 120: 173-184.

MOHN, R.K. 1999. Simulation studies of causes of retrospective patterns in analytical assessments. Working Paper for ICES Comprehensive Fisheries Evaluation Working Group. January, 1999. 27 pp.

NRC 1998. Improving Fish Stock Assessments. Report of the Committee on Fish Stock Assessment Methods. National Research Council, National Academy Press. Washington, DC., 177 pp.

PARSONS, L.S. 1993. Management of Marine Fisheries in Canada. Can. Bull. Fish. Aquat. Sci., 255: 763 pp.

PEPPER, D. 1998. What are the fish for? Allocation and Evaluation in the Salmon Fishery of BC. West Coast Fisherman, 13: 21-26.

RICE, J.C. (in press). A review of the trajectories of Canadian Atlantic cod stocks 1977-1997, and past explanations of the trajectories. Can. J. Fish. Aquat. Sci Spec. Publ., 22 pp.

RICE, J.C. & RICHARDS, L.J. 1996. A Framework for Reducing Implementation Uncertainty in Fisheries Management. N.A. Jour. Fish. Manag., 16: 488-494.

RICKER, W.E. 1975. Computation and Interpretation of Biological Statistics of Fish Populations. Bull. Fis. Res. Board Can., 191: 382 pp.

SPISAK, R. 1998. Fishermen's Forum: Reaction to the 1998 Salmon Management Plan. West Coast Fisherman, 13: 19-23.

STEPHENSON, R. 1998. Trawling for Bureaucrats and the Hake Joint Venture Fishery. West Coast Fisherman, 13: 15-18.

STEPHENSON, R.L. & LANE, D.E. 1995. Fisheries Management Science: A plea for conceptual change. Can. J. Fish. Aquat. Sci., 52: 2051-2056.

WALTERS, C. 1986. Adaptive management of renewable resources. (Macmillan Publishing, New York, N.Y., U.S.A., p.384

WALTERS, C & MAGUIRE, J.-J. 1996. Lessons for stock assessment from the northern cod collapse. Rev. Fish Biol. Fish., 6: 125-137.

Bycatch control through technical regulations and fisheries monitoring

Åsmund Bjordal

Institute of Marine Research, P.O. Box 1870, N-5024 Bergen, Norway.
Email: [email protected]

Abstract: The current Norwegian fisheries management regime has been developed since the early 1970s. With a major goal to obtain sustainability in the fishing industry, the exploitation of most commercial fish stocks are now limited by total allowable catch (TAC) regulations and minimum legal fish sizes for most important species. Additionally, the management strategy includes different approaches to minimise various types of bycatch. Adequate bycatch control is attempted through different management measures such as minimum legal fish size, temporary closure of fishing grounds and a general discard ban. To obtain the desired effects of these regulations, emphasis is put on control and enforcement at sea as well as control of the landings. The increased complexity of different regulatory measures has, however, complicated fishing operations. From the fishermen's point of view, the price of sustainability is therefore, a complicated situation on the fishing grounds, with reduced efficiency and short-term profitability. To minimise this conflict of interests, research on improved size- and species selectivity of different fishing gears has been emphasised. The work has particularly been focused on improved selectivity of trawl gear, where the bycatch problems have been most pronounced. In addition to codend mesh regulations, the development of sorting grids in different trawl fisheries have been proven to be successful solutions to minimising bycatches of unwanted sizes and species. Sorting grids have also been developed for size grading in various purse seine fisheries. Technical solutions for improved selectivity of stationary fishing gears have also been developed, e.g. different approaches to reduce the bycatch of birds in longline fishing. Technical measures and devices for sorting fish during the capture process have limited value if this induces high mortality on the released organisms. Results from studies of by-mortality related to different bycatch reducing devices are therefore, described and discussed.





Figure 1. (Clockwise) a) Temporary closed areas for demersal fish trawling in the Barents Sea in 1998, b) Areas closed for demersal fish trawling in the Barents Sea, from 24 March 1998, c) Areas closed for demersal fish trawling in the Barents Sea, from June 1998, d) Areas closed for demersal fish trawling in the Barents Sea, from 1 October 1998, e) Areas closed for demersal fish trawling in the Barents Sea, from November 1998. (By J. P. Hansen, Norwegian Directorate of Fisheries).


The development of size sorting devices in groundfish trawls in order to reduce the bycatch of undersized fish was achieved through the development of another type of grid, termed the Sort-X system (Larsen & Isaksen 1993) (Fig. 3). This system is essentially a reversed shrimp grid. As fish travel back through the trawl they make contact with the grid with large fish forced backwards and downwards along the sloping grid to the cod end. Fish that are small enough pass through the slots in the grid and escape from the trawl through an open panel in the top surface of the net. From 1997 grid-sorting devices were made mandatory in all Barents Sea groundfish trawl fisheries. In Norwegian trawls the Sort-X system is currently in use and a simpler system (Sort-V) is used in Russian trawls (see Fig. 3). The Sort-V system which is easier to handle and less expensive to construct has now been successfully adapted to Norwegian trawls and will most probably be legalised for use in the Norwegian fishery in the near future.

Recent trials with sorting grids in other trawl fisheries have also recorded promising results. These have been particularly encouraging in the size selection of mackerel (Scomber scomber) by the pelagic trawling fleet and in the separation of cod and haddock in the North sea trawl fishery for Norway pout (Trisopterus esmarkii) and blue whiting (Micromesistius poutassou) (Kvalsvik et al. 1998, Huse et al. 1998).

In mixed species fisheries, fishermen often experience bycatch problems when legally sized fish of a filled quota are taken. Research on methods for species selective devices in groundfish trawls has therefore, been the focus of recent research and a promising solution for the separation of cod from saithe and haddock has been developed (Fig. 4). Behavioural observation of the main commercial species indicated that where cod tended to swim downwards, saithe, haddock and other species swam upwards when encountering trawl gear. Based on these observations a selection device was developed which divided the trawl into an upper and lower part using a horizontal, longitudinal panel. Field trials gave very promising results with the majority of the cod captured in the lower cod end and the majority of the haddock and saithe taken in the upper cod end. This solution could allow fishermen an opportunity to continue fishing a cod quota in a mixed fishery even if the haddock quota is reached by letting the haddock escape through an open upper cod end (Engås et al. 1998).

Figure 2. The Nordmøre grid, sorting grid for shrimp trawls. (From Isaksen et al. 1992).

Figure 3. Size sorting grids for groundfish ; a) Sort-X (Larsen and Isaksen 1993), b) Sort-V (Lisovsky et al. 1996).

Figure 4. Groundfish trawl with horizontal panel for species separation (Engås et al. 1998).

Grids have also been tried in seine nets to select and separate fish by size. Although the results have been promising, the handling of grids with this gear has been adversely cumbersome. The use of square mesh cod ends has, however, significantly improved selectivity in the seine net fishery for cod and haddock, and is now being used on a voluntarily basis by many fishermen (Fig. 5). Grids have also been tried to sort fish by size in different purse seine fisheries and good results have been obtained, particularly with saithe and mackerel (Fig. 6).

A different bycatch problem that has received the focus of research attention in recent years has been the capture of seabirds by longline gear. During the setting of longlines, seabirds are attracted to the baited hooks that are accessible in the setting zone directly behind the vessel. Different solutions have been suggested and encouraging results have been obtained using either a setting funnel that guides the line below the diving depth of the scavenging birds or a bird scaring device consisting of a line with vertically hanging streamers that is trailed above the critical zone behind the vessel (Bjordal and Løkkeborg 1996). Recent experiments have shown that the seabird scaring device seems to be a superior solution to the problem with a significant reduction in the seabird bycatch and improved catch rates of the target fish species recorded when this device is deployed successfully (Løkkeborg, 1998).

In longline fishing it has been demonstrated that the choice of bait can have clear species-selective effects which can be utilised to minimise unwanted bycatch of certain species. An example is the use of a recently developed restructured bait in the mixed fishery for cod and haddock. Compared with traditional bait, the new bait catches 2-3 times more haddock, but it gives reduced catches of cod. In situations with restricted cod quotas and more liberal haddock quotas, longline fishermen have used the new bait to minimise the catch of cod. The fishermen have, therefore, been able to prolong the fishing period in this mixed fishery which would otherwise have been closed once the cod quota had been taken.


Figure 5. Seine net. Cod-ends with diamond (top) and square (bottom) meshes. (Isaksen et al. 1997).

Figure 6. Size sorting grid in a purse seine.

Figure 7. Scaring device for the avoidance of seabird bycatch in longlining. (Bjordal and Løkkeborg 1996).



BJORDAL, Å. & LØKKEBORG, S. 1996. Longlining. Fishing News Books, London, 156 pp.

CHOPIN, F.S. & ARIMOTO, T. 1995 The condition of fish escaping from fishing gears - a review. Fish. Res. 21: 315-327.

ENGÅS, A., JØRGENSEN, T. & WEST, C.W. 1998. A species-selective trawl for demersal gadoid fisheries. ICES J. Mar. Sci. 55: 835-845.

HUSE, I., KVALSVIK, K., GAMST, K., FOSSEIDENGEN, J.E., GODØY, H. & TORGENSEN, Ø. 1998. Selective fishing for industrial fish. (Cruise report R/V "Johan Hjort")

ISAKSEN, B., GAMST, K. & MISUND, R. 1997. Comparison of handling and selectivity properties of grids and square mesh codends in the seine net fishery. Fisken Havet (In press).

ISAKSEN, B., LØBACH, T. & VEIM A.K. (submitted). The Norwegian management system: an example of a necessary prerequisite for development, implementation and acceptance of selectivity devices in fishing gear. ICES Mar. Sci. Symp., 000: 000-000.

ISAKSEN, B., VALDEMARSEN, J.W., LARSEN, R.B. & KARLSEN, L. 1992. Reduction of fish by catch in shrimp trawl using a rigid separator grid in the aft belly. Fishing Research 13: 335-352.

KVALSVIK, K., MISUND, O.A. & GAMST, K.A. 1998: Selectivity experiments with grids in a mackerel trawl, December 1998. Internal report, Institute of Marine Research, Bergen, Norway. (in Norwegian).

LARSEN, R.B. & ISAKSEN, B. 1993. Size selectivity of rigid sorting grids in bottom trawls for Atlantic Cod (Gadus morhua) and haddock (Melanogrammus aeglefinnus). ICES Mar Sci. Symp. No 196: 178-182.

LISOVSKY, S.F., SAKNOE, V.A. & KONDRATSJUK, 1996. Recommendation for using the Sort-v grid system based on a single-grid during trawl fishery on Arcto Norwegian Cod. Murmansk PINRO 14 pp. (In Russian).

LØKKEBORG, S. 1998. Seabird by-catch and bait loss in longlining using different setting methods. ICES Journal of Marine Science 55: 145-149.

The role of fisheries monitoring programmes in identifying and reducing problematic bycatches

Steven J. Kennelly

New South Wales Fisheries Research Institute, P.O. Box 21, Cronulla, NSW 2230, Australia.
Current address: Centre for Research on Ecological Impacts of Coastal Cities, Marine Ecology Laboratories, A11 University of Sydney, NSW 2006, Australia.
Email: [email protected]

Abstract: The first pre-requisite for any attempt to reduce unwanted bycatches in fisheries is accurate information on the species, quantities, sizes, locations and timing of such bycatches. Such information not only facilitates the identification of any spatial and temporal closures to fishing designed to reduce bycatches, but also allows fishing gear technologists to develop modifications that reduce bycatches whilst maintaining catches of the targeted species. There are several methods available to quantify bycatches and discards (e.g. questionnaires, interviews, logbooks, samples from fishers, data from research vessels), but it is well-accepted that the most accurate way to estimate bycatches is by using onboard observers. Observer programmes involve having fishery-independent scientists or observers gathering data during the course of normal fishing operations. If the survey design, sampling frequency and extent of the observer programme is adequate, the data gathered can be used to estimate species- and size-specific bycatches by the whole fishery across the spatial and temporal scales required for subsequent bycatch reduction programmes.


Table 1.Summary of the main characteristics of the 3 main ways used to identify and quantify bycatches and discards.

Category Examples Costs to managing agency Inconvenience to industry Precision Accuracy Reliability Representation of normal fishing
Fishery-dependent surveys Interviews with fishers, logbooks, samples collected by fishers. Low High Low Low Low High
Fishery-independent surveys Data from research vessels and chartered commercial vessels. High None High High High Low
Observer programmes Fishery-independent observers on normal fishing operations Intermediate Intermediate High High High High








ALVERSON, D.L., FREEBERG, M.H., MURAWSKI, S.A. & POPE, J.G. 1994. A global assessment of fisheries bycatch and discards. FAO, Fish. Tech. Paper No. 339, FAO, Rome, 233 pp.

BROADHURST, M.K. & KENNELLY, S.J. 1994. Reducing the by-catch of juvenile fish (mulloway Argyrosomus hololepidotus) using square-mesh panels in codends in the Hawkesbury River prawn-trawl fishery. Fish. Res., 19: 321-331.

BROADHURST, M.K. &. KENNELLY, S.J. 1995. A trouser-trawl experiment to assess codends that exclude juvenile fish (mulloway) in the Hawkesbury River prawn-trawl fishery. Mar. Freshwater Res., 46: 953-958.

BROADHURST, M.K. & KENNELLY, S.J. 1996a. Effects of the circumference of codends and a new design of square-mesh panel in reducing unwanted by-catch in the New South Wales oceanic prawn trawl fishery, Australia. Fish. Res., 27: 203-214.

BROADHURST, M.K. & KENNELLY, S.J. 1996b. Rigid and flexible separator-panels in trawls that reduce the bycatch of small fish in the Clarence River prawn-trawl fishery, Australia. Mar. Freshwater Res., 47: 991-998.

BROADHURST, M.K. & KENNELLY, S.J. 1997. The composite square-mesh panel: a modification to codends for reducing unwanted bycatch and increasing catches of prawns throughout the New South Wales oceanic prawn-trawl fishery. Fish. Bull., 95: 653-664.

BROADHURST, M.K., KENNELLY, S.J. & ISAKSEN, B. 1996a. Assessments of modified codends that reduce the by-catch of fish in two estuarine prawn-trawl fisheries in New South Wales, Australia. Fish. Res., 27: 89-111.

BROADHURST, M.K., KENNELLY, S.J. & O'DOHERTY, G. 1996b. Effects of square-mesh panels in codends and of haulback-delay on bycatch reduction in the oceanic prawn-trawl fishery of New South Wales, Australia. Fish. Bull., 94: 412-422.

BROADHURST, M.K., KENNELLY, S.J., WATSON, J. & WORKMAN, I. 1997. Evaluations of the Nordmøre-grid and secondary by-catch reducing devices (BRDs) in the Hunter River prawn-trawl fishery, Australia. US Fish. Bull., 95: 210-219.

DANNEVIG, H.C. 1904. Preliminary report upon the prawning industry in Port Jackson. W.A. Gullick, New South Wales Government Printer, New South Wales Government, Australia, 17 pp.

FAO 1996. Report of the Technical Consultation on Reduction of Wastage in Fisheries. Tokyo, Japan, November, 1996. FAO Fisheries Report No. 547, FAO, Rome, 27pp.

GAUVIN, J.R., HAFLINGER, K. & NERINI, M., 1995. Implementation of a voluntary bycatch avoidance program in the flatfish fisheries of the Eastern Bering Sea. Solving By-catch: Considerations for Today and Tomorrow. Alaska Sea Grant College Program Report No. 96-03, University of Alaska Fairbanks, pp. 79-85. ISBN 1-56612-038-1.

GRAY, C.A., MCDONALL, V.C. & REID, D.D. 1990. By-catch from prawn trawling in the Hawksbury River, New South Wales: species composition, distribution and abundance. Austr. J. Mar. Freshwater Res., 41: 13-26.

GUTHERZ, E.J. & PELLEGRIN, G.J. 1988. Estimate of the catch of red snapper, Lutjanus campechanus, by shrimp trawlers in the U.S. Gulf of Mexico. Mar. Fish. Rev., 50: 17-25.

HOWELL, W.H. & LANGAN, R. 1987. Commercial trawler discards of four flounder species in the Gulf of Maine. N. Amer. J. Fish. Man., 7: 6-17.

HUDON, C. 1990. Distribution of shrimp and fish by-catch assemblages in the Canadian eastern Arctic in relation to water circulation. Can. J. Fish. Aquat. Sci., 47: 1710-1723.

ISAKSEN, B., VALDERMARSEN, J.W., LARSEN, R.B. & KARLSEN, L. 1992. Reduction of fish by-catch in shrimp trawl using a rigid separator grid in the aft belly. Fish. Res., 13: 335-352.

JERMYN, A.S. & ROBB. A.P. 1981. Review of the cod, haddock, and whiting discarded in the North Sea by Scottish fishing vessels for the period 1975-1980. International Council for the Exploration of the Sea CM 1981/G:47.

KENNELLY, S.J., 1995. The issue of by-catch in Australia's demersal trawl fisheries. Rev. Fish Biol. Fish., 5: 213-234.

KENNELLY, S.J. & BROADHURST, M.K. 1996. Fishermen and scientists solving by-catch problems: examples from Australia and possibilities for New England. Solving By-catch: Considerations for Today and Tomorrow. Alaska Sea Grant College Program Report No. 96-03, University of Alaska Fairbanks, pp. 121-128. ISBN 1-56612-038-1.

KENNELLY, S.J., LIGGINS, G.W. & BROADHURST, M.K. 1998. Retained and discarded by-catch from oceanic prawn trawling in New South Wales, Australia. Fish. Res., 36: 217-236.

LIGGINS, G.W. & KENNELLY, S.J. 1996. By-catch from prawn trawling in the Clarence River estuary, New South Wales, Australia. Fish. Res., 25: 347-367.

LIGGINS, G.W., KENNELLY, S.J. & BROADHURST, M.K. 1996. Observer-based survey of by-catch from prawn trawling in Botany Bay and Port Jackson, New South Wales. Mar.Freshwater Res., 47: 877-888.

MEDINA, H. 1994. Reducing by-catch through gear modifications: the experience of the tuna-dolphin fishery. In: By-catch in fisheries and their impact on the ecosystem. University of British Columbia Fisheries Centre Research Report 1994 Vol. 2 No. 1. ISSN 1198-6727, p 60.

SAILA, S.B., 1983. Importance and assessment of discards in commercial fisheries. FAO Fisheries Circular No. 765, 62 pp.

TILLMAN, M.F., 1993. Bycatch - The issue of the 90's, p. 13-18. In: R.P. JONES, ed., International Conference on Shrimp By-catch, May, 1992, Lake Buena Vista, Florida. Southeastern Fisheries Association, Tallahassee, Florida.

WATSON, R.A., DREDGE, M.L.C. & MAYER, D.G. 1990. Spatial and seasonal variation in demersal trawl fauna associated with a prawn fishery on the central Great Barrier Reef, Australia. Austr. J. Mar. Freshwater Res., 41: 65-77.

Fisheries monitoring: A proposed Canadian model for an "Integrated conservation and management system" (ICAMS)

David Bevan1 and Dennis Brock2

1 Director General, Conservation and Protection Directorate, Department of Fisheries and Oceans Ottawa, Ontario, Canada.
2 Director of Enforcement, Conservation and Protection Directorate, Department of Fisheries and Oceans Ottawa, Ontario, Canada.

Abstract: The ICAMS approach is a logical business approach to the orderly management of a highly regulated industry. The focus is on industry monitoring and verification of the processes that are used to ensure the sustainability of the fishery resources. The regulatory agency retains legislative control therefore, ensuring compliance.


1.1 Background

1.2 Extension of jurisdiction

1.3 Integrated fish management plans

  1. Recognition that the primary objective for the conservation of fishery resources be that of sustainable utilisation by Canadians;
  2. The establishment of a partnering initiative to foster a more innovative and self reliant fishing industry;
  3. The establishment of an Integrated Fisheries Management Planning (IFMP) process with all DFO sectors and clients, including fish harvesters, providing input to the development of management plans, and
  4. The development and implementation of the conservation and protection (enforcement) programme renewal strategy.

1.4 Current challenge


2.1 Application of risk management

2.2 Changing responsibilities and roles


Table 1. A comparison of hazard analysis critical control point (HACCP) and integrated conservation and management system (ICAMS) principles.




Principle 1

Identify the potential hazards

Conduct analysis of the fishery.

Principle 2

Determine points, procedures and operational steps.

Identify conservation and management threats, action points and measures.

Principle 3

Establish critical limits associated with each hazard.

Establish critical limits associated with each conservation and management threat.

Principle 4

Establish a monitoring system.

Establish the monitoring system.

Principle 5

Establish the corrective action.

Establish corrective action plan.

Principle 6

Establish procedures for Verification.

Establish verification processes.

Principle 7

Establish documentation Procedures.

Auditing and Enforcement actions.

Principle 8


Documentation and recording keeping.

3.1 Application to fisheries management


4.1 Conduct an analysis of the fishery

  1. Define the scope of the fisheries management plan;
  2. Collect all relevant information regarding the species (intended use, participants, fleet sectors, processors, other), stakeholders (gear types, fishing areas and seasons, history of landing and values) and all other relevant aspects of the fishery;
  3. Describe and evaluate the effectiveness of previously applied conservation and management measures, and threats;
  4. Assemble a working group/advisory committee, and
  5. Establish the work plan for the development and implementation of the Integrated Fishery Management Plan (IFMP).

4.2 Identify the conservation and management threats, action points and measures ( Annex II)

  1. The science programme should be responsible for the initial description of the conservation threats, the most appropriate action points and the conservation measures that may be applied;
  2. The descriptions of the conservation threats, action points and measures should be included with the stock assessment and biological advice;
  3. Identify the management threats, action points and measures;
  4. Fishery management should be responsible for the initial description of the management hazards, action points and measures;
  5. Conservation and management threats, action points and measures should be confirmed by the working group/advisory committee;
  6. Identify, with the working group/advisory committee, the most appropriate fisheries conservation/management measures and techniques that will be used to minimise the risk associated with each conservation and management threat, and
  7. Identify the action points where the conservation or management measures would be most effectively applied.

4.3 Establish critical limits associated with each conservation and management threat.

  1. Prioritise the conservation and management threats;
  2. Determine the level of risk associated with each threat i.e. the potential impact to conservation of the stock and management of the fishery (high, moderate, low);
  3. Determine the level of risk associated with each of the proposed fishery conservation and management measures (high, moderate, low);
  4. Identify the critical limits associated with each preventative conservation and management threat and measure;
  5. Describe the application of the conservation and management measures in the format of a fishery management plan, and
  6. Establish the requirements for the development of compliance-related activities; monitoring, verification and audit and enforcement of the "Integrated Fishery Management Plan".

4.4 Establish the monitoring system.

  1. Identify the level of monitoring required to achieve the critical limits of each threat and measure at each CAP and MAP;
  2. Determine the most effective monitoring activity and frequency required to achieve compliance at each CAP and MAP;
  3. Determine who in the fishing enterprise will have responsibility for monitoring the conservation and management measures at each CAP and MAP. For example, a specified crew member, and
  4. Recognise that the license holder is at all times accountable for the monitoring process.

4.5 The corrective action plan

4.6 Establish the verification process

4.6.1 Industry directed verification

4.6.2 Shared verification

4.6.3 Government directed verification

4.7 Auditing and enforcement actions

4.8 Documentation and recording keeping

4.9 Terminology


Table 2. The roles and responsibilities of the regulatory agency and the fishing enterprises in the fisheries management process.

Role Current Responsibility Responsibility Under ICAMS
Enacting Legislation and Regulations



Setting Public and Fisheries Policy



Establishing Management Plans Government/Industry Government/Industry
Monitoring Government/Industry Government/Industry
Verification Government Government/Industry
Auditing/Enforcement Government Government

Assignment of the monitoring role to industry would permit the regulatory agency to direct its resources to auditing functions. In addition, the regulatory agency could direct additional resources to specific conservation threats and measures that could not be assigned to industry because of the sensitivity of the threat and the impact it could have on the sustainability of the fishery.

Internally, the roles of the various DFO programmes may not require significant modification and the current roles have been designed to compliment the IFMP process. The provision of advice and the development of the IFMPs would remain the essential roles of the science and fisheries management programmes.

The most significant changes would occur in the responsibilities of each programme (Table 3). The ICAMS approach requires that the individual programmes complete specific components of the IFMP. For example, science could be responsible for articulating the conservation threats, management measures and action points. Fisheries management could articulate the management threats, management measures and action points (Table 2). The process of developing the IFMP in terms of activities i.e. analysis, advice and consultation, is compared in the context of the existing approach and ICAMS. Both processes result in the completion of the IFMP. However, the ICAMS approach clearly, describes and prioritises the conservation and management threats, management measures and action points. This level of detail in the IFMP is essential to the effective development and implementation of the monitoring, corrective action, verification, audit and documentation processes.

The focus of ICAMS is directed primarily on those actions that are included in monitoring, control and surveillance activities. However, the shift of responsibility to the fishing enterprises for the monitoring function requires increased emphasis on the verification and audit functions of government.


Table 3. The development of an integrated conservation and management system (ICAMS) approach from existing integrated fisheries management plans (IFMP) and management activities.

IFMP Guidelines Management Activity ICAMS Principles
Overview of Fishery Analysis Conduct Analysis
Stock Status Scientific Advice Identify Management
Threats & Measures
Management Advice

General Management Objectives

Management Advice


Identify Management Threats & Measures


Current Management Issues Consultation Establish Critical Limits
Management Issues Fishery Management Plan Proposed Fishery Management Plan
Enforcement Issues & Strategies Monitoring Control and Surveillance Monitoring System

Corrective Action Plan

Verification Process

Auditing and Enforcement



  1. Provides for access to the decision-making process through the existing "Integrated Fish Management Plan" process;
  2. Provides for a consistent and rigorous decision-making process that can address all the conservation and management concerns in integrated fish management plans;
  3. Describes the principles that provide a consistent framework and methodology for decision making and establishing priorities for fishing plans;
  4. Provides for an analytical framework for the implementation of multi-year fishery management plans;
  5. Provides for a framework for the effective involvement of the harvesting sector in all aspects of the "Integrated Fish Management Plan" process;
  6. Provides for a framework and methodology for the harvesting sector to develop monitoring and verification processes and strategies for inclusion in its conservation harvesting plans;
  7. Upon adoption is comparable to the establishment of a quality assurance/control process for fisheries management. This has an impact similar to introducing a total quality assurance programme;
  8. Provides an opportunity to direct resources toward those measures that have the potential for greatest impact on the sustainability of the fishery, allowing maximum utilisation of existing resources in consequence;
  9. Provides for an increase in the responsibility and accountability of all players in the IFMP process;
  10. Provides for an effective due diligence defence if a significant event, such as collapse of a fish stock, were to occur through the rigorous application of the standards and principles contained in the ICAMS approach, and
  11. Compliments the development and implementation of the "Code of Conduct for Responsible Fishing".


CASHIN, R. 1993. Charting a new course: towards the fishery of the future. Report of the taskforce on incomes and adjustment in the Atlantic fishery. Communications Directorate, Department of Fisheries and Oceans, Ottawa, Ontario, Canada.

CODEX ALIMENTARIUS 1993. Guidelines for the Application of the hazard Analysis Critical Control Point System. ALINORM 93/131, Appendix II.

DILLON, M. & GRIFFITH, C. 1997. How to Audit: Verifying Food Control Systems. M.D. Associates.

DILLON, M. & GRIFFITH, C. 1995. How to HACCP; An Illustrated Guide. M.D. Associates.

EMBERLEY, B. J. & ROWE, L. W. 1998. An Integrated Conservation and Management System, A concept Paper. Department of Fisheries and Oceans, Ottawa, Ontario, Canada.

FISHERIES RESOURCE CONSERVATION COUNCIL 1997. A Groundfish Conservation Framework for Atlantic Canada. Report to the Minister of Fisheries and Oceans. Department of Fisheries and Oceans, Ottawa, Ontario, Canada.

Fish Inspection, Quality Control and HACCP, A Global Focus. Proceedings of the Conference held May 19-24 1996, Arlington, Virginia, USA.

MCEACHERN, V. Canadian Food Inspection Agency, Ottawa, Ontario. Personal communication.

Previous PageTop Of PageNext Page