P A R T   II
REPORT OF THE WORKING GROUP ON RESOURCES STUDY AND MONITORING

by

A.D. MacCall (Chairman)*

* A.D. MacCall is with the National Fisheries Service, c/o Southwest Fisheries Centre, P.O. Box 271, La Jolla, CA 92038, U.S.A.

1. INTRODUCTION

To be fully effective, biological monitoring and studies require consideration of fishery context and alternative methods. This report examines some of these considerations, and reviews some recent developments. This discussion is intended as a supplement to the many manuals presently circulating. Recommendations are difficult to make outside the environmental, biological, and socio-economic context of particular fisheries. In some cases, general needs are clear. For the remaining cases, this document provides information which may be useful in evaluating needs and choosing between alternative methods and studies.

This document is the product of the Working Group on Resources Study and Monitoring which met during the Expert Consultation on Neritic Resources. The composition of the Working Group varied during the Consultation. In addition to their technical input, R. Jones and I. Tsukayama provided valuable service as rapporteurs. Many other people participated regularly in the Working Group and/or provided substantial sections of this document. These included J. Alheit, B. Brown, J. Carscadden, R. Crawford, P. Fréon, M.L. Garcia, J. Lleonart, J. Lopez, A. Menz, S. Saccardo, H. Santander, R. Serra, G.D. Sharp, P. Shelton, E. Ursin and many more.

1.1 Working criteria

Fishery science stems recently from the much older European tradition we call “science”. This tradition focusses almost exclusively on the academic pursuit of knowledge and the establishment of truth. However, fishery science extends beyond that tradition in that it forms a working basis for altering our world; that is, it is an applied science. The consequences of this aspect are not commonly recognised in fishery work, and in many respects call for a different approach to that which is taught in the university.

Table 1. Comparative time scales

In addition to the traditional scientific criterion of knowledge, there are other criteria which should guide our efforts. Because the results of our work may contribute to changes in society as well as ecosystems, fishery scientists bear a burden of responsibility which is generally lacking in normal academic science. An important aspect of this criterion is timeliness. Many fishery analyses and recommendations are specific to the current status of the fishery, and lose value if they are not communicated quickly. The next criterion is closely related, and will be called appropriateness. This criterion is particularly important to the task of this Consultation; recommendations must not only reflect existing knowledge and responsible interpretation/application, they must also be suited to the needs of the recipient.

Fishery science is the application of ecological principles (in the broadest sense) to particular fisheries. Thus fishery models specifically, and fishery science generally, must be viewed situationally like situational ethics in the philosophical sense; any technique or measurement should not be viewed in the abstract, but in relation to the particular fishery under consideration. Unfortunately, this document, being a general review, necessarily violates the last criterion.

1.2 A note on terminology

The use of broad terms like pelagic or demersal when referring to fish groups causes a dichotomy that does not exist in nature. For example, pelagics are said to be highly variable, but haddock on Georges Bank has been more variable than herring in the Gulf of Maine, and mackerel in the recruited ages has a lower natural mortality rate than cod. Therefore fishery scientists should be specific in discussing these groupings and use species group designations (i.e. sardine/anchovy) rather than terms like pelagic. The characteristics upon which the groups are based should be explicitly stated.

2. A PERSPECTIVE OF VARIABILITY

Variability of fishery resources occurs on nearly all time scales. Moreover, biological variability is strongly influenced by physical or environmental variability on one side, and variability in the human sector on the other side (Table 1). Much of the dynamics of fisheries and the difficulties in their exploitation and management histories (e.g. booms and collapses) can be inferred from the interaction of forces acting on different time scales. As in the classical ecology of predator-prey cycles, delays in fishery exploitation response to resource abundance creates an inherently cyclic or unstable system. By recognising the nature of these interactions and designing appropriate and effective regulations, management can control and minimize these technological fishery cycles, within the bounds of natural environmental variability.

3. CAUSES OF CHANGES IN FISH ABUNDANCE

Fishery and resource monitoring often covers a wide selection of biological variables. This activity is done with the implicit assumption that changes in monitored variables provides information on the status of the resource, with emphasis on productivity and abundance. In order to evaluate the utility of fishery monitoring activities, it is useful to more explicitly examine the relationship between biological variables and causes of change in fish abundance.

Causes of change in fish abundance fall into five categories:

• Intraspecific dynamics, which includes compensatory mechanisms such as stock-recruitment relationships, density-dependent growth and cannibalism.

• Competition among species, which is of considerable interest but in practice is difficult to demonstrate conclusively.

• Predation, which is generally treated as natural mortality in fishery analyses.

• Fishing, or exploitation

• Environmental fluctuations, particularly abiotic factors and lower trophic level biological factors.

Table 2. Monitored biological variables that provide evidence for causes of variability in abundance of Engraulis. Symbols qualify the relationship (XX = strong evidence; X = evidence; i = indirect connection; ? suspected)

¹ Examples are guano production or seabird reproductive success
² This includes mixed schooling and geographic/temporal overlap

Biological variables, if properly monitored, can provide evidence of the causes of change listed above. Table 2 lists the most commonly monitored biological variables and the related causes of variability for the anchovy (Engraulis) fisheries. Nearly all variables are related to environmentally caused variation, and very many are related (at least indirectly) to fishing. Competition is a possibility in many cases, but no variable provides substantial evidence for competition. Intraspecific mechanisms are mostly evidenced by direct study of the fish, including its early life stages. Notably, the latter are some of the few variables not related to fishing. Some biological variables provide little useful evidence for causes of variability, for example, sex ratio is commonly monitored but is unlikely to provide useful evidence for causes of variability.

Table 2 would be different for other species of fish. Because Engraulis has been studied more than most species, the strength of evidence for other species will tend to be lower. In some cases additional biological variables would be appropriate. For example, northwest Atlantic herring have historically been impacted by disease, suggesting that monitoring for this factor would have some value.

4. SYMPTOMS OF ADVERSE CHANGES IN RESOURCE STATUS

Fishery stock assessment attempts to evaluate current stock productivity in relation to its potential productivity. This potential is estimated from historical resource monitoring, biological studies, and comparative information from similar species or fisheries. Experience has shown that fish stocks respond to exploitation in predictable ways, but environmental fluctuations often complicate the patterns. Because of this fundamental similarity among fishery responses, several models have been developed which concisely summarise the relationship between the fishery harvest and the resource. For example, as intensity of fishery removals increases, abundance decreases, mean age or size of fish decreases, and age at first maturity decreases. These changes in themselves simply reflect fishing pressure and compensatory responses. The level at which they reflect poor stock condition (i.e. performance falling below potential, due to excessive fishing) requires consideration of the context of historical levels and fluctuations as well as other features of the resource which may also have changed. For this reason, a generalised discussion of indicators of symptoms of adverse changes in stock condition is of very limited utility. However, for the purpose of discussion, Table 3 lists some common symptoms and indicators, their most likely interpretation, and some of the considerations that need to be taken into account.

Before concluding that a symptom signifies an adverse condition, the nature of the fishery and resource must be understood. Exploitation risk varies substantially with species and environmental characteristics, as shown in Table 4. With this background, the symptom in Table 3 must be examined for possible alternative causes such as environmental changes or changes in market demand. Once the symptom is accepted as a valid indicator of stock condition, the severity of the symptom must be judged on an objective basis. Usually this consists of using historical information in a fishery model which has well-established methods of interpretation. The model may also indicate the nature and extent of remedial action necessary to rehabilitate the stock.

The categories of diagnostic symptoms are few - relating to abundance, recruitment (incoming young fish), and relative intensity of harvesting. In addition, there are a variety of warning signs that are more difficult to interpret. Abundance is usually the most important indicator of stock condition, and for this reason a substantial portion of the research and monitoring effort should be devoted to this aspect. If a direct estimate of abundance is not available there may be other indicators such as the catch rate of the fishery, the geographic extent of the stock, or the status of more visible stock-dependent predators such as seabirds. Due to the relative imprecision of these auxiliary indicators, confirmation from more than one source is desirable. Analytical techniques such as production modelling may provide an estimate of “healthy” abundance levels (e.g. the level of abundance giving maximum productivity). Depending on the natural variability of the resource, a “warning” level of abundance might be half of the abundance giving maximum average productivity. A “danger” level could be similarly defined at a somewhat lower level.

Table 3. Most Common Signs of Deteriorating Resource Status and Potential Problems

Table 4. Comparison of “Contexts” for six Selected Fisheries (from Beverton, 1983)

Recruitment is the main source of fish biomass which replaces losses from the stock due to harvest and natural mortality. The higher the total death rate, the more sensitive is the resource to recruitment variability and/or failure. Recruitment fluctuations due to environmental variability are mostly unavoidable, although many resources show smaller relative recruitment fluctuations at higher parental stock sizes. When recruitment declines in parallel with decreasing parental stock abundance, there is a high potential for depletion (“recruitment overfishing”). When a history of recruitments is available, a plot of recruitment versus parental stock, with or without a fitted regression line, is useful to identify not only the average relationship, but often more importantly, to identify the existence of runs of above - or below - normal recruitment. Experience in Peru and Namibia have shown that anomalous fat or oil content patterns provide early warning of anomalous ocean conditions (e.g. El Niño) and related likelihood of future recruitment failure.

The activity of harvesting impacts a resource mainly due to the additional mortality (fishing mortality) imposed. There seems to be a consistent relationship between the magnitude of the natural mortality rate (M) and sustainable levels of fishing mortality rate (F). The relationship is the basis of the popular rule-of-thumb F_opt ≃ M, but fishery experience has shown this often to be an overestimate of optimal F. Thus, a symptom of overfishing is a fishing mortality rate which approaches the natural mortality rate in magnitude. An alternative indicator is when mean age or length falls near the age or length at first spawning. Besides being evidence of relatively high fishing pressure, this condition indicates imminent recruitment decline due to lack of spawning potential.

Several other stock conditions can be taken as warning signs, but do not provide specific diagnoses. For example, increasing fluctuations in catch suggest a possible loss of compensatory capacity, as well as increased risk of collapse. Similarly, deviations from normal physiological or behavioural patterns may be due to environmental fluctuation or may be normal responses to fishing pressure, but should not be dismissed without consideration of possible deterioration of stock condition.

Importantly, Table 3 is not exhaustive. There are many more possible symptoms and indicators, some of which may have general utility, and many of which have specific utility in particular fisheries. Once again, it must be stressed that every fishery is in some respects unique, and the relevant symptoms and considerations vary accordingly.