S. Garcia, Fishery Resources Division, FAO, Rome. Italy.
Paper presented at the 2nd World Fisheries Congress. Workshop on Fisheries
Sustainability Indicators, Brisbane, Australia, August, 1996
Following the work of the World Commission on Sustainable Development and the Brundlandt Report, the issue of using information for more informed planning and decision making has been central to the development debate for all sectors, and particularly for those exploiting natural renewable resources. Following UNCED (Rio de Janeiro, 1992) the issue has gained further momentum and most sectoral and inter-sectoral organizations (including non-governmental ones), policy-making structures and management bodies are addressing the issue (in particular, the Scientific Committee on Problems of the Environment (SCOPE) in cooperation with UNEP). This involves assessing the state of the resources (often bad), addressing social and economic constraints (generally serious) and institutional failure (resulting from ineffective laws and organizations), and discovering the relative inadequacy of the information base and analytical capacity available to support decision making. Information available is rarely sufficient for fully informed decision making and simple and carefully selected indicators could improve the effectiveness of decision processes.
There are hundreds of definitions of sustainable development, leading potentially to a wide range of diverging criteria to define sustainability indicators or to interpret their variations. For fisheries, a useful definition would be the one agreed at FAO in 1988 which states that sustainable development has been defined by the World Commission on Environment and Development (1987) as "development that meets the needs of the present generation without compromising the ability of future generations to meet their own needs". It has also been defined by the FAO Council in 1988 as "the management and conservation of the natural resource base, and the orientation of technological and institutional change in such a manner as to ensure the attainment of continued satisfaction of human needs for present and future generations. Such sustainable development conserves (land,) water, plants and (animal) genetic resources, is environmentally non-degrading, technologically appropriate, economically viable and socially acceptable". This definition implies an objective of optimizing welfare from a limited natural resource base, minimizing resource and environmental degradation, and regulating the rate of use of these resources over time. "Welfare" is defined by Shun and Archibald (1996) as including "the value of natural amenities, improvement in environmental quality, reduction of pollution and waste, and value of the inter-generational equity". Pontecorvo and Schranck (1995) refer to "the welfare goal of maximizing the long-run net economic yield from the resource base" which would "eliminate short-run profit maximization.. and incorporates the biological goal of conservation".
Dahl (1996) lists some of the many dimensions of sustainability (fiscal, ecological, human, social, moral, ethical, and spiritual), each of which could, potentially, lead to development of indicators.
The concept of sustainability has been embedded in fisheries literature for about half a century at least in the concept of maximum sustainable yield (MSY). This concept has been used for decades as a measure of fishery potential and, unfortunately, sometimes as a development and management target. In the post UNCED era, the general concept requires explicitly that both the conditions of the ecosystem and the people living in it be either "good" or improving. There may be a number of interpretations of "goodness". Prescott-Allen (1996) considers that "ecosystem well-being is a condition in which ecosystems maintain their quality and diversity and thus their potential to adapt to change and provide a wide range of choices and opportunities. Human well-being is a condition in which all members of the society are able to define and meet their needs and have a large range of choices and opportunities to fulfil their potential. A sustainable society would be able to achieve both conditions as well as (have) the capacity to anticipate change and recover from eventual setbacks".
THE NEED FOR SUSTAINABILITY INDICATORS
Chapter 40 of Agenda 21 calls for the development of indicators of sustainable development and the first and second sessions of the Conference on Sustainable Development (1993 and 1994) stressed the need for a "Menu of Indicators" as a basis for: (a) establishing cost-effective data collection systems; (b) monitoring conditions and trends in the fishery sector; (c) informed decision making; and (d) as a possible basis for early warning systems.
Sustainability indicators (SIs) are pointers which would be used to reveal and monitor the conditions and trends in the fishery sector. They would allow monitoring the sustainability of the fishery sector and the fishery development policy and management performance in relation to the various components of the fishery system: the environment, the target resource, the associated and dependent species, the economic and social conditions, and the cultural context. Ideally, SIs should look at environmental, resource, economic and social elements of sustainability in an integrated manner.
Indicators could be needed, for instance, by a State to judge whether the owners of exclusive fishing rights are complying with the duty of care imposed on them, or whether a straddling stock or highly migratory resource is exploited in compliance with the Law of the Sea and other relevant international instruments (such as the FAO Code of Conduct on Responsible Fishing). They could also be used by NGOs and the public at large to assess the performance of a national policy or management system. In international trade, and following recent developments, they could also be used as a basis for an eco-certification and labelling system of fisheries and fishery products as proposed by the World Wide Fund for Nature (WWF) and an important worldwide fish trader (Unilever) in the framework of a Marine Stewardship Council.
Indicators could also be used dynamically to compare the trajectory of a fishery with a planned (target) trajectory. For instance, they could be used to compare the evolution of spawning biomass during the implementation phase of a resource rehabilitation strategy with the planned trajectory, taking corrective decisions when the indicator diverges from the target trajectory. Dahl (1996) gives examples of such use for population and pollution indicators.
Many fisheries are studied using complex bio-eco-sociological models of growing complexity and very demanding data requirements. Their results are often very complex and their presentation may vary greatly between models. These results need to be presented to all stakeholders in a simple, readable way. In addition, the information required for complex models is not always available with low data requirements. There is therefore a need for a simple and robust system for tracking instability, through the use of a limited number of indicators with some integrative properties (i.e., the fluctuations of which reflect changes in more than one component of the fishery sector).
The need for indicators creates, in turn, a need to (a) develop more accurate information systems to be updated in quasi real time; (b) develop indicators sufficiently specific to be of practical value for management, and sufficiently integrated and generic to be comparable across fisheries; (c) to integrate the environmental, biological, economic and social indicators into aggregate ones. The development of indicators therefore poses a significant challenge to countries and particularly to their research institutions.
CRITERIA FOR SUSTAINABILITY INDICATORS
In order to be acceptable and effective, indicators and criteria for sustainable development should be relevant, reflecting key forces and properties of the exploited ecosystem as well as sufficiently accurate and precise. To be used for management, they should be based on a large consensus among interested parties, particularly but not only on international fisheries. Such consensus is usually obtained through guiding and binding agreements and conventions, as well as agreed standards and methodological protocols. In this post-UNCED era, and considering the requirement for precaution enshrined in UNCED Principle 15, indicators of sustainability should explicitly take account of the uncertainty in the data and estimation procedures. While general rules may apply, they should also be context-sensitive and adapted to the area and the species concerned. Finally, they should also relate explicitly to management objectives and constraints as reflected in management reference points. These considerations are further developed below.
Relevance, scope, timeliness, accuracy and precision
The prime qualities of any type of information for decision making are: relevance, scope, timeliness, accuracy and precision. It may seem evident that a sustainability indicator should be of direct relevance to the issue of sustainability. It may be less evident that an indicator may not be relevant in isolation and that a combination is often required to ensure such relevance. Total catches, for instance, while generally available, give little clue as to whether the fishery is sustainable or not in the absence of complementary data on fishing effort and species composition. Similarly, data on total fleet capacity (in terms of number of canoes or total Gross Registered Tonnage (GRT) or horsepower) may not really reflect the trend in fishing pressure in the absence of data on the operation of the fleet, particularly its rate of use. Because sustainability is a complex issue involving a number of considerations related to the ecosystem and society, a system of indicators is needed, the scope of which should cover both the resources to be used sustainably and the goods and services obtained from the system as well as relevant societal parameters. In addition, indicators should preferably be unbiased and their trends should really reflect the evolution of the magnitudes they are supposed to represent. They should be estimated with sufficient precision to allow distinction between the main "signal" the indicator is supposed to provide and the residual "noise" related to error in the data or in the system representation (e.g., the assumed cause-effect relationships).
Degree of consensus
As indicators are aiming, inter alia, at institutionalizing transparency, they require agreement among those for whose use they are intended. In general, "sustainability" is perceived differently by different users. At national level, indicators of sustainability might be selected and imposed by the government on the basis of its own criteria and objectives. If decisions affecting wealth and its distribution will be made on the basis of these indicators, governments will have to develop consensus around them between ministers and user-groups. A fortiori, indicators need to be agreed upon in international exploitation systems for shared, straddling and highly migratory stock. This may become necessary if procedures for eco-labelling and eco-certification of products or management systems are put in place. This implies the existence of an international mechanism to develop such agreement as well as legal texts, agreements and guidelines around which consensus can be developed based on scientific analysis. The existence of a regional fishery body or management arrangement is a pre-requisite. In all cases, an interactive process of development of agreed objectives would greatly contribute to the process of definition of sustainability indicators.
There are a number of important internationally agreed instruments, the provisions of which may or should be implemented in developing and managing fisheries. The central ones are the 1982 UN Convention on the Law of the Sea (UNCLOS) which came into force in 1994, the 1995 UN Agreement for the Effective Implementation of the 1982 Law of the Sea Convention in relation to Straddling Fish Stocks and Highly Migratory Fish Stocks (hereafter called the 1995 UN Agreement), and the 1995 FAO voluntary Code of Conduct for Responsible Fisheries. There are also a number of agreements related to, for example, the marine environment, or to shipping, which might lend themselves to the development of indicators of interest to fisheries.
In order to conform with UNCLOS sustainability indicators should be based on "the best scientific information available". This means that the indicators should: (a) be based on accessible and verifiable information, and (b) calculated from a documented and peer reviewed protocol. This also implies that MSY "as qualified by relevant and economic factors" (UNCLOS Article 61) should be used as a benchmark (but not as a development target). In many instances, in order to facilitate comparative approaches among fisheries and areas and in order to ensure stability of the indicator with time, conventions might need to be developed agreeing formally on the data to be used, the criteria to be followed, and the derivation methodology, establishing the indicators as international standards.
To facilitate agreement, selected indicators should be accompanied with detailed information concerning: (a) type of indicator: pressure, state, or response indicator; (a) purpose of the indicator; (b) relevance to policy; (c) relevance to sustainable development; (d) linkages with other indicators; (e) targets; (f) related international conventions and agreements: identifying the commitment; (g) data requirements; and (h) appropriate methodologies.
Precautionary character
Recognizing the level of uncertainty in the scientific understanding and data, sustainability indicators should also reflect a precautionary approach to fisheries as required by UNCED (Principle 15), and foreseen in the 1995 UN Agreement and the 1995 FAO Code of Conduct. Indicators should be developed taking into account the available guidelines for the Precautionary Approach to Fisheries (Garcia, 1994; FAO, 1995, 1996). This implies that the indicators and criteria should be selected and estimated in a way that explicitly takes into account their uncertainty (e.g., by estimating their confidence intervals, using precautionary criteria), and minimizes risk for the resource and the people. Precaution should also be reflected in the management set-up, the characteristics of which could be used as "institutional" indicators of sustainability (see the section below on "Indicators of structure"). For instance, a precautionary management strategy should incorporate continuing collection of appropriate data and have sufficient flexibility to allow a quick effective reaction to any signs that something is going wrong. A fishery solely regulated through total allowable catches with no control on fishing capacity would be less precautionary than a fishery based on strict controls in capacity and quantitative fishing rights. The subject is elaborated further below.
Specificity
While a number of potential indicators are probably of general use, particular characteristics of species, stocks or fishing areas affecting resilience of resources or fisheries sustainability should be taken into account as well. This could be done by using different indicators for different types of resources (e.g., birth rate for marine mammals, spawning biomass for sardines) as well as different criteria to indicate the boundaries within which the indicator is allowed to fluctuate (e.g., the minimum allowable reproduction potential would be set at a higher level for whales than for sardines). This latter device would be useful in increasing the level of precaution for particularly sensitive, endangered or unstable species. Indicators should also take account of the local socio-economic conditions and traditions.
Linkage with management objectives and reference points
Fisheries scientists and management have been using indicators and criteria of sustainability for a long time even though the terminology was not in general use. For example, catch rates, stock biomass, recruitment levels, costs, revenues, etc. are traditionally calculated (and often standardized) in fishery science and can be related to sustainability. There are many possible sources and types of indicators in fisheries, but the interpretation of their fluctuations in terms of sustainability will depend totally on the criteria used to set allowable limits to these fluctuations.
In seeking criteria for sustainability of a fishery, it is essential to examine the management objectives which have been designed for sustainability. These are the first place in which the principles for sustainable fisheries should be reflected, demonstrating at least the intention of meeting sustainability requirements. Making a judgement on objectives requires that they are explicitly stated, and in a sufficiently quantitative manner to allow performance evaluation. In practice, explicit statement of objectives are rare and in most cases they are expressed as "motherhood" statements, not permitting any objective assessment except perhaps on the long-term and in retrospect.
Fishery management objectives indicate the limit towards which the fishery is aimed and indicators can be developed (and usually are) to monitor the performance of the fishery in relation to them. However, these could only be considered as "sustainability" indicators if the objectives themselves have been selected with sustainability in mind. For example, the evolution of annual catches in relation to a fixed total allowable catch, or to MSY, would not tell much about the sustainability of the fishery in the absence of effort data. In addition to production targets, fisheries should have conservation objectives, expressed as targets or constraints, and the indicators should show the state of the fishery in relation to them as well as the rate at which they are evolving towards or away from them. Conservation targets exist when management aims at them (e.g., within a rebuilding strategy). A conservation constraint exists when, for example, a minimum biological limit (considered "safe" for the resource) is set. In this case indicators should show that the fishery does not cross this limit.
Management reference points are of general use in fisheries to reflect objectives (Target Reference points, TRPs) or constraints (Limit Reference Points, LRPs, and Threshold Reference Points, ThRPs) and would be essential criteria for sustainability, delimiting the allowable area of fluctuation of the indicators. The use of management reference points and their role in relation to precaution are discussed in Caddy and Mahon (1995), Garcia (1994, 1996), and FAO (1995).
NATURE AND TYPES OF INDICATORS
Fishery sustainability indicators (FSIs) should be selected for their ability to indicate whether a fishery is sustainable (responsible) or not. However, there are many requirements for sustainability and therefore many potential indices of sustainability based on the structure of the fishery system and its performance in relation to agreed criteria.
Weak and strong sustainability
In selecting indicators and establishing the criteria by reference to which performance will be assessed, it may be convenient to refer to definitions of sustainability as a starting point. There are many definitions, some of which differ fundamentally, for example the concepts of "weak" or "strong" sustainability. Weak sustainability allows full substitution among all forms of capital (natural, economic and social) and would allow depletion of the natural capital provided the sum of all three is kept constant for future generations or increases over time. Strong sustainability, on the contrary, assumes that forms of capital are not interchangeable and should be conserved separately. This implies that fisheries resources should not be depleted beyond their natural capacity of renewal and passed as such to future generations. This is in line with the concept of sustainable yield enshrined in UNCLOS and indicates that, for fisheries, indicators should reflect and aim at a form of strong sustainability. In practice, a compromise between the two is chosen based on the concept of acceptable levels of impact.
A system of indicators
The FAO Council definition of sustainable development given in the introduction might be an acceptable starting point to identify issues and indicators. The definition refers to the need to control:
1. the resource base;
2. technological change; and
3. institutional change.
It further stipulates that such controls are needed to ensure:
1. satisfaction of human needs for present and future generations;
2. conservation of land, water, plants and animal genetic resources;
3. non-degradation of the environment;
4. use of appropriate technology;
5. economically viable exploitation; and
6. socially acceptable situations.
Based on this definition alone (and there are many others), indicators, each of which may integrate more than one variable, would be needed to track:
1. the resource endowment, including its abundance, diversity and resilience;
2. the environment, for example by reference to its pristine condition;
3. the technology in terms of capacity as well as environmental-friendliness;
4. the institutions, e.g., fishing rights, enforcement system;
5. the human benefits, e.g., food, employment, income;
6. the economics of exploitation, e.g., costs, revenues, prices;
7. the social context, e.g., social cohesion, participation, compliance.
Indicators of pressure, state, and response
There are many potential indicators that could contribute to track sustainability and, in order to avoid dispersion and overloading of the information collection system, a small number of them will need to be selected. The selection should be issue driven, focusing efforts on:
1. indicators of pressures (direct and indirect) or driving forces affecting the resource system;2. indicators of the state of the system being affected; and
3. indicators of response reflecting actions taken (by management, or industry, or other stakeholders) to mitigate, reduce, eliminate, or compensate for the stress. Such action could be taken to affect the pressure (mitigation, regulation) or the state (compensation, rehabilitation).
These three types of indicators characterize the pressure/state/response framework (PSR) 1 for which examples are given in Table 1.
1 Sometimes also called the driving force/state/response framework.
TABLE 1. Examples of indicators of pressure, state and response
|
Issue |
Pressure (driving force) |
State (Condition) |
Response (Mitigating action) |
|
Overfishing Economic losses |
Overcapacity |
Biomass < MSY |
Limit access |
|
Littoral habitat degradation |
Coastal trawling |
% seagrass cover |
Protected areas |
|
Extensive aquaculture (and other pressures) |
% of mangrove cover |
Mangrove replanting |
|
|
Algal blooms |
Pollution |
Nutrient load |
Aquaculture feed - management |
1 Land-based sources of pollution.
When considering indicators (of pressure, state or response) it will be essential to consider the linkages between them (e.g. between two pressure indicators or between a pressure and a response indicator) as well as the response time (or inertia) of the system, i.e., the time required for a pressure or a response to be fully reflected in the state. Both relate to the dynamics of the resource and the exploited sub-systems and their components. They do not seem to be well represented in the PSR concept which seems to relate better to the long-term e.g. when addressing long standing issues, in which pressures have been exerted for a long time, leading to a chronic "state of nature" reflecting a degree of dynamic equilibrium between pressures, responses, and state 1. The PSR concept captures neither the direction of the trend in the pressure and state and their rate of change, which are fundamental, nor the lag time between the moment the pressure is applied (or changed) and the moment at which the effect is fully reflected in the state.
1 In complex bio-socio-economic systems, it is unreasonable to assume "equilibrium" sensu stricto.
In practice, it may not always be simple to distinguish between indicators of "state" and indicators of "pressure" and some indicators could be considered either way depending on the particular point of view. Catches, for instance, are an indicator of the level of extraction (a pressure) as well as of the resource (a state). Salaries could be taken as reflecting a state of human well-being or a pressure inasmuch as they could reflect the incentive to fish. Another example is given by the demand for fish. Its increase is a sign of human well-being as well as a potential signal for increased pressure to fish.
Indicators of level, change and structure
The situation of an exploitation system in terms of pressure, state or response can be more comprehensively described by indicators of level, change and structure.
Indicators of level reflect the spatial or temporal evolution of key system variables expressed as absolute values (e.g. fishing capacity, spawning biomass, revenues, employment, number and gravity of conflicts) or in the form of ratios (e.g., between virgin and present biomass, or between fishery and agricultural revenues). They measure the final response of the system or of one of its components. They integrate a large number of interactions and reflect directly the performance of the system, if used in conjunction with peer-established sustainability criteria. However, indicators of level (e.g., stock abundance), in isolation, may not provide information on whether the system is stable, improving or worsening, nor on the action eventually required (e.g. on Monitoring, Control and Surveillance (MCS)? On licensing system? On capacity controls? On market prices or taxes?). Types of such potential indicators are listed in Appendix 2. The example in Figure 1 illustrates the theoretical changes in spawning biomass under fishing pressure, decreasing from the virgin level (Bv) to the lowest allowable limit (Blim) and below, to raise again through management (e.g., effort reduction) to the target level fixed in a management plan. The figure illustrates the fact that the situation requires attention as soon as biomass decreases and becomes unacceptable, requiring corrective action, when biomass is driven below the agreed limit.
FIGURE 1. Theoretical representation of a level indicator (of spawning biomass) and its fluctuations under varying fishing pressure in relation to limit and target levels. Bv = virgin biomass, Blim = minimum acceptable (safe) limit.
Some indicators may not be unequivocally interpreted, however. For instance, high rates of legal action and conviction in a fishery could indicate an active enforcement system which is an element favourable to sustainability. However, they could also indicate a high level of non-compliance which is a serious sign of non-sustainability even in its weak form.
Indicators of change indicate the direction and rate of change of key indicators. Combined with indicators of level, they give a dynamic perspective to otherwise static indicators and would be particularly needed for early warning systems. Dahl (1996) refers to "vector indicators, showing the direction of speed and movement towards or away from a goal".
Indicators of structure refer to the functional elements of the system. When referring specifically to institutions they have been called institutional indicators. They could be considered as conditions which are necessary for sustainability but which, per se may not be sufficient to guarantee it, the final outcome depending on effectiveness of implementation. Nonetheless, the sustainable (responsible) nature of an exploitation system, management setup, or development strategy can be assessed against a checklist of desirable and undesirable system properties relating, inter alia, to: (1) the objectives retained for development or management; (2) the management planning process with its institutions, and mechanisms; (3) the management approach and measures; and (4) the management implementation.
Regarding the objectives which should be clearly specified, sustainability should be the long-term overriding objective in a set that should address minimum requirements for resource and environment conservation as well as sectoral production targets. For example, a fishery system aiming at taking two-thirds of the maximum sustainable yield (2/3 MSY) or the maximum economic yield (MEY) would, in principle, be more sustainable than a system aiming at taking the full MSY with the increased risk for the resource that this objective entails.
Regarding the management planning process and institutions, elements contributing to sustainability are, for example: (1) the ongoing collection and availability of data and indicators regarding the potential, present state and trends in the environment, the resources, the various fishery sub-sectors; (2) the existence of a coastal zone planning and management system explicitly including fisheries; (3) the establishment of mechanisms to ensure effective people's participation; (4) the existence of dispute resolution mechanisms and qualified voting procedures; (5) the allocation of property or use rights; (6) mechanisms for allocation of wealth; (7) existence of economic disincentives to effectively control overcapacity; (7) the setup of a high-level independent committee to advise on the development policy in relation to long-term sustainability and oversee its implementation; and (8) a minimum level of fishery research that can support preparation and follow-up of development and management plans.
Regarding management approaches and measures, positive signs would include: (1) existence of effort reduction processes (e.g. buy-back schemes) and access regulations; (2) payment for fishing rights commensurate with the value of the resource; (3) regulations on discarding practices, protection of juveniles; (4) establishment of marine protected areas as a means to conserve biodiversity; (5) non-availability or phasing out of subsidies; (6) existence of macro-economic instruments as incentives or disincentives.
Regarding management implementation, positive signs would include: (1) a credible control and surveillance system (satellite monitoring, observers on board); (2) an effective system of collection of catch and effort statistics; (3) a streamlined judicial process with deterrent penalties; (4) an in-built process of periodic evaluation of fisheries and management performance coupled with rolling development and management planning processes.
While the above examples referred essentially to the development and management planning processes, other parameters of importance relate to the sector itself, for example: (1) its degree of organization and participation in decision making; (2) its efforts to raise awareness among its people; (3) the quality of the statistics it submits; (4) its cooperation in enforcement.
A much more comprehensive record of structural and process requirements for sustainability is available in the FAO Code of Conduct for Responsible Fisheries (FAO, 1996) which, as an agreed international instrument, and together with its guidelines for practical implementation, can be taken as a basis for assessment of fisheries potential sustainability.
In most cases, these institutional indicators could be considered as "sustainability switches" which can be either "ON" or "OFF". In some instances, however, they might lead to the establishment of a proper level indicator. For example, the ON-OFF switch on "participation mechanisms" could perhaps be turned into an indicator of the level of participation as suggested below (with arbitrary gradations and value judgements):
0.0-0.19 = Bad = top-down, non participative system, mere provision of data
0.2-0.39 = Poor = people consulted after decisions are made, for comments
0.4-0.59 = Medium = people participate in the debate
0.6-0.79 = OK = people share some decision power
0.8-1.00 = Good = people are involved, including in enforcement.
A matrix of indicators
As an example of the typology given above, a system of indicators in relation to a particular issue (overfishing) would fit into the matrix given in Table 2.
TABLE 2. An indicator matrix
|
Indicators |
Pressure (Capacity) |
State (Biomass) |
Response (Management) |
|
Level |
C > Cmsy |
B < Bmsy |
Effort reduction scheme |
|
Change |
C + 5%/year |
B - 4%/year |
C - 5%/year for 5 years |
|
structure |
3 interacting fleets |
|
Fishing rights |
INTEGRATION OF INDICATORS
Appendix 1 lists many potential candidates for sustainability indicators, of unequal importance, with non-quantified and sometimes unknown inter-relationships. There would be an obvious advantage to reduce the list to a manageable minimum. It would probably be optimistic, however, to hope that some single indicator could be used to track sustainability. In practice, a number of carefully selected indicators might be needed on the sustainability "dashboard".
Just as objectives, indicators may refer to bio-ecological, environmental, social and economic components. It may be possible and it would indeed be useful and necessary to identify partially integrated indicators, reflecting overlapping components of sustainability (e.g., eco-biological and techno-economic or socio-cultural. Figure 2). A more difficult challenge will be the development of fully integrated indicators, the changes of which would capture the changes of sustainability itself.
It would be useful to elaborate a framework that accommodates the full range of social, environmental and economic factors of the sustainability nexus. The Human Development Index of UNDP is an example of a complex indicator combining per caput GDP, adult literacy, and life expectancy appropriately weighted to give an indicator of living standards. The task appears very difficult however. Fisheries have used sustainability criteria for decades. International agreement was achieved on MSY, its properties and estimation protocols, despite its shortcomings as a criterion, and MSY was enshrined in the 1982 Convention on the Law of the Sea. However, the concept of optimum yield (OY), which was intended to integrate in addition, environment and socio-economic considerations, was never used to any great extent as no agreement could be reached on a universal protocol to estimate it. Instead, the F0.1 target reference point (Gulland and Boerema, 1973) has found widespread application because of its simple definition and internationally agreed conventional estimation procedure 1. This reference point corresponds to lower fishing pressure and costs, and higher spawning biomass and profits than FMSY, at a small cost in terms of employment, bringing the fishery close to an otherwise undefined "optimum".
1 F0.1 corresponds to the level of fishing mortality (or fishing effort) at which the marginal yield (i.e., the increase in yield obtained by one additional unit of effort) is 10% of the marginal yield at origin (i.e., observed at extremely low levels of fishing).
This would indicate that there are potentially three ways towards integration of sustainable indicators:
1. determining separate indicators related to resource, environment, economic and social concerns, integrating them in a reference system as proposed for instance by Prescott-Allen (1996) and as illustrated below;2. formally adopting a single indicator, based on one easily measured variable with recognized "integrative" properties (such as F0.1), as a proxy for an "integrated" or "partially integrated" indicator; and
3. a combination of the two.
SUSTAINABILITY REFERENCE SYSTEMS (SRSs)
The sustainability barometer of Prescott-Allen (1996)
Prescott-Allen (1996) has proposed a "sustainability barometer" based on a graphical representation of the location of an exploited ecosystem on an orthogonal system in which the two axes represent indexes of human well-being and of ecosystem well-being, considered as the two fundamental dimensions of sustainability (Figure 3). The aim of the barometer is to (a) give a picture of the whole system; (b) treat ecosystem and human well-being as equally important; (c) facilitate a rigorous and transparent progress towards sustainability. Used as orthogonal axes, the human and ecological dimensions, with a scale normalized between 0 and 1, provide an orthogonal system of reference in which the position of an exploitation system (e.g., a fishery) can be located if the corresponding values on the two axes can be estimated.
The scales of the barometer include also "value judgements" corresponding to the various intervals on the axes, e.g. the 0.0-0.2 interval is considered "Bad" while the 0.8-1.0 interval is considered "Good". Prescott-Allen stresses the importance of the "scaling" of the barometer and the amount of case-specific judgement involved in it. The paper does not explain how the numerical value of the coordinates is arrived at but examples are given in this paper in the specific case of fisheries. Prescott-Allen called it a "sustainability barometer" used to "measure" exploitation pressure, by analogy with the instrument used to measure atmospheric pressure. Because this device does not provide a "measure" of sustainability but helps representing it, locating an exploited ecosystem in a system of reference, in the rest of this paper I shall refer to it and to other similar devices as "Sustainability Reference Systems" (SRSs).
Figure 3 gives a representation of Prescott-Allen's concept, slightly modified to indicate not only the qualitative sustainability scale but also the areas of social or ecological instability. The conditions expressed in the matrix and the implications in terms of ecosystem or human well-being refer to the long term. It is recognized that, in the short term, a degradation of the ecosystem may help improve people's well-being. It is also recognized that improving the ecosystem for future generations may require a short-term, temporary decrease in human well-being.
Prescott-Allen indicates that the two scales of the SRS are independent reflecting the fact that the reference system is not a correlation matrix, and stresses that improving one set of conditions should not lead to a degradation of the other - a principle frequently violated in practice. For the same reason, and by design, lower scores always override higher scores (e.g., Good X Bad = Bad) to show that there can be no long-term compromise with bad conditions. This is because bad or poor conditions of either the ecosystem or humans are not considered sustainable, and the condition recognized for the system, by convention, will be the condition of the worse dimension. It is recognized that trade-offs exist between people and is the ecosystem in the short term. In the long term, however, if one dimension is damaged, the whole system is damaged and "there can be no trade-offs between people and the ecosystem". On Figure 3, the X and Y scales are identical i.e., the categories "good", "poor", etc., correspond to the same gradation and type of scale, leading to a symmetric SRS. This may, however, not always be the case. One could imagine that scales could be different, i.e., the same category "good" could cover a different range of the numeric scale on each axis of the SRS, and the scales themselves could be arithmetic or logarithmic, for example.
The concept expressed in Figure 3 is static and falls short of capturing the dynamics of the fishery system. A complete assessment of the degree of sustainability of a fishery should take into account both the static and dynamic aspects of the situation, i.e., position of the fishery system on the SRS, and the local and global trend in such a position, i.e., whether the situation is improving or worsening. It is clear, for instance, that no matter what the static coordinates are, simultaneous decreases in the conditions of ecosystem and humans are a sign of instability and will not be sustainable.
The dynamics of the system could be captured in two ways which could eventually be combined. First of all, once the methodology to calculate the two references (human and ecological) is available, and data permitting, the position could be calculated in retrospect for many years in the past. The sequence of positions on the SRS would "map" the trajectory of the fishery and allow some trend extrapolation. Similarly, if a rate of change (e.g., annual rate) could be estimated in the two dimensions, the resultant of the simultaneous changes would give an indication of the likely future position of the fishery.
Figure 4 illustrates this additional concept. Assuming that a fishery could be located on a SRS, the direction in which (and the rate at which) the situation is changing would be as important as the position on the SRS. Direction and rate of change would indeed provide useful foresight. The combination of Figure 3 and 4 (on Figure 5) would lead to a representation of the fishery system on the system of reference, at the intersection of its aggregated coordinates on the ecosystem and human conditions axes. The dynamics would be captured by the observed trajectory and two arrows, the size and direction of which would reflect the rate of change and its sign (positive or negative). Figure 5 illustrates the importance of adding dynamism to the SRS and clearly shows that for the same position on the SRS, different types of action may be required to correct the trajectory.
It would not be easy in most cases to represent the complexity of the fisheries sustainability equation in a referential system with only two axes, and a more complicated system could be imagined as described below.
A sustainability kite diagram
Star diagrams are often used to represent multivariate properties of a system, e.g., to summarize the performance of a computer with scores referring to its performance in terms of processor velocity, RAM capacity, hard disk capacity, file transfer speed, energy efficiency, interface user-friendliness, etc. Figure 6 gives a theoretical example of such a diagram and illustrates the fact that it can be used to compare the profile (the "signature") of different systems including the "ideal" one with optimal values for all parameters.
FIGURE 6. Theoretical example of a star diagram
A theoretical example of such a diagram for fisheries, using only 4 axes (kite diagram) for the sake of simplicity, is given in Figure 7. The parameters represented are arranged in two domains corresponding respectively to ecosystem and human well-being (in order to remain in the terminology used by Prescott-Allen (1996)). Each axis can be scaled from 0 to 1 and the grey scale refers to the assessment categories used in the preceding SRS (black = Bad, light grey = Good). A fishery can be re-presented on this referential system by a polygon and two fisheries can be compared by comparing their polygons. In addition, the position of the polygon in relation to each axis indicates in which sphere action might be required to improve the situation.
Scaling the axes of Sustainability Reference Systems
Prescott-Allen gives a detailed account of many of the problems encountered and options available when scaling the axes of the SRS. Scaling requires the determination of the scale boundaries (0-1 or 0-100) and the relevant subdivisions of that scale according to the value judgements (e.g., deciding whether "Bad" goes from 0 to 0.2 or from 0 to 0.5). The latter could sometimes be arbitrary or conventional, but should in most instances refer to the target and limit reference points. In the example given by Prescott-Allen for the sustainability barometer the two axes are scaled from 0 to 1 and the value judgements (i.e., Good to Bad) are evenly distributed on both axes (cf. Figure 3). In most instances, the true values of the sustainability indicators (e.g., the size of the spawning biomass) will not be between 0 and 1 but, say, between the value of Bv, the biomass of the virgin stock, and zero. In this case, re-scaling will be needed, e.g., by using ratio indicators (e.g., B/Bv). In the section on "Indicators of level", above, an attempt has been made to scale, from 0 to 1, the degree of people's participation in a management system and arbitrary value judgements were given. To use the SRS, the same effort would be required for all potentially useful indicators, using as quantitative methods as possible for the estimates, and a set of criteria for the value judgements. Some examples are given in Table 3 exclusively for the purpose of illustration.
TABLE 3. Examples of scaling of indicators
|
Scale |
State |
Pressure |
Pressure |
Response |
|
Good |
0.5 - 1.0 |
0.6 - 0.8 |
0.8 - 1.0 |
0.8 - 1.0 |
|
OK |
0.3 - 0.5 |
0.0 - 0.6 |
0.5 - 0.8 |
0.6 - 0.8 |
|
|
|
0.8 - 1.0 |
1.0 - 1.2 |
0.4 - 0.6 |
|
Medium |
0.2 - 0.3 |
1.0 - 1.3 |
1.2 - 1.4 |
|
|
Poor |
0.1 - 0.2 |
1.3 - 2.0 |
1.4 - 2.0 |
0.2 - 0.4 |
|
Bad |
0.0 - 0.1 |
> 2.0 |
> 2.0 |
0.0 - 0.2 |
1: Assuming a Limit Reference Point at 30% Bv and a Target Reference Point at 50% Bv2: Assuming a Target Reference Point at F = 60 to 80 % of FMSY
3: Assuming a Target Reference Point at Economic Yield (EY) = 80-100% of Maximum Economic yield (MEY)
The examples given show that the scale may not always be between 0 and 1 and the scaling of the value judgements may also vary between indicators. This last point is illustrated in Figure 8. The value judgements "Good", "Poor", etc. related to the "Participation" indicator, for instance, have been arbitrarily established. Where feasible, and as shown for the other examples given in the table, it would be preferable to relate them to the management reference points corresponding to the desirable (= Good) or undesirable(= Bad) states of the system components as objectively determined on biological or socio-economic grounds. To reflect consensus, particularly in international fisheries, the scaling of value judgements would have to be agreed among interested parties.
FIGURE 8. Example of different scaling of value judgements for different elements of the system
A theoretical but practical example of scaling is given in Figure 9, using two simplistic relationships well known to all fishery biologists and economists: the relationship between a pressure (fishing effort) and two resulting states (the biomass and the economic rent). This figure has the merit of illustrating the fact that:
1. The functions to be re-scaled may be linear or non-linear.2. Despite the assumption of independence between the two axes of the SRS, the indicators used may indeed be dynamically and functionally correlated. Action to displace one indicator may also displace the other, sometimes in an unpredictable manner.
3. Because of non-linearity, a value on one axis may correspond to more than one value on the other, with different implications in terms of sustainability and of action required.
4. The reference to standard criteria such as MSY or MEY or BMSY for the value judgements may often lead to dissimilar value judgement scales even though the numerical scale could be identical.
When the axes of the SRS are identical both quantitatively (e.g., both are from 0 to 1) and qualitatively (e.g., on both scales the value judgements correspond to the same range) the SRS could be called "isometric" (as in Figure 7). However, when different scales and particularly different value judgements are used (as in Figure 10) the SRS is "anisometric". Considering that the shading pattern (i.e., the position of the value judgements on the axes) reflects the policy framework (i.e., the set of target and limit management reference points) the anisometric SRS shows explicitly the direction in which the policy is oriented and, in particular, how precautionary it is. It allows therefore for a comparison between policy frameworks as well as an assessment of the performance of particular fisheries - by the position of the fishery kite on the SRS. A comparison between Figures 7 and 10 illustrates the fact that the same fishery kite (i.e., the same set of values for a fishery system) leads to different diagnostics. While in Figure 7 the fishery kite is entirely situated in the Medium-OK area of the SRS, the same fishery appears "Bad" in relation to its ecological axes, in Figure 10, reflecting the fact that, in that SRS, the ecological limits have been set higher.
Scaling and weighting indicators
Combining indicators requires weighting them according to their importance for sustainable development (as agreed by the interested parties). If indicators are averaged into a single value as suggested by Prescott-Allen, weighting is essential and would reflect policy direction by comparison of the weights given to "conflicting" indicators (e.g., of human or ecosystem well-being). In Figure 9, where indicators need not be averaged, it is the scaling of the value judgements which can be taken as reflecting the weight attached to each indicator.
Scaling the SRS and target setting
In the above examples, I have explicitly related the scaling to target and limit reference points. Prescott-Allen underlines that scaling and target setting are two different things and that best values on the scale are not necessarily targets. This is particularly obvious for spawning biomass, for instance, where the "best" value for the stock would most likely be the virgin one (implying no exploitation) while the "best" for fisheries would be some value close to, but higher than the biomass at MSY.
FIGURE 11. Stochastic and dynamic representation of a sustainability kite.
SRS and dynamics of the system
The process of identifying and codifying integrated indicators and of their representation in an SRS will require exploring the relationships between the key variables of the exploitation system. The example given in Figure 9 showed the integration in a SRS of an indicator of ecosystem well-being (the spawning biomass of the target species) and of human well-being (the aggregated economic rent extracted). The figure shows that each indicator is a function of fishing effort (one linear and one non-linear) and that the values of the two indicators are related (even though the SRS is not designed to reproduce this relation). This example shows that:
1. taking action to change one of the indicators will have an impact on the other;2. one level of human well-being (rent) may correspond to more than one value of the ecosystem well-being (two in the example given) indicating an economic alternative. It is recognized that the alternative also has different implications for human well-being, in terms of employment for instance.
The 4-axis representations given in Figures 7 and 10 are deterministic and static. The addition of confidence limits of the kite and of indicators of change at each of its angles (Figure 11) would give a more accurate and dynamic picture of the situation, indicating the most likely extent and direction of the changes. The ends of the vectors would in fact define the forecast kite. If the indicators of change would be made to reflect the expected impact of governmental policies, the forecast kite would represents a target against which achievements could be compared later on to judge the policy performance, and SRS would be integrated in the management implementation and evaluation cycle.
CONCLUSION AND DISCUSSION
The biological components of sustainability have been familiar to fishery scientists and managers for some time, and are enshrined in the 1982 UNCLOS concept of maximum sustainable yield which must be explicitly related to the conditions of the environmental, resource, economic, social and institutional aspects of fisheries. Despite the obvious complexity of the sustainability issue, the number of parameters that condition it, and the range of potentially useful indicators, there is a need for a simple representation of sustainability by indicators, as integrated as possible. These indicators are intended to complement the sophisticated bio-economic simulations used traditionally in fisheries as a basis for the analysis of management options. They offer a simple way to integrate social considerations, and a generalized representation, explicitly related to sustainability, and allowing comparative analysis and easy access to the information by a large non-technical audience.
Sustainability indicators are also needed to increase the number of system variables effectively used in management and to promote the development of more performance - sensitive management systems than those presently in place - which have demonstrated their lack of reaction to clear signs of failure.
The development of sustainability indicators requires: (1) consensus among interested parties; (2) reference to agreed sets of principles, rules and concepts: (3) standard protocols for their calculation, based on accepted, peer-reviewed scientific methodologies and "the best scientific information available". Indicators should be accompanied by detailed information related to: (1) type (pressure, state or response indicator); (2) purpose; (3) relevance to policy; (4) relevance to sustainable development; (5) linkages with other indicators; (6) targets; (7) relations with international conventions and agreements; (8) data requirements; and (9) appropriate (recommended) methodology.
In fisheries, several important international agreements are available from which the requirements for indicators could be extracted (a task that this paper has not attempted). In this respect, and to underline only one important aspect of these requirements, indicators should reflect a precautionary approach to fisheries and be customized to take account of the nature of the resource, environment or human communities involved. They should also explicitly relate to management objectives and reference points.
Indicators should reflect the "strong sustainability" concept internationally agreed for marine fisheries and enshrined in UNCLOS (with little practical implementation until now!). The indicators on the management "dashboard" should relate to the pressures exerted on the system, its state (i.e., the values of its key variables) and the management response using indicators of level, change and structure. Indicators should provide information on the bio-ecological, environmental, social, and economic components of the fishery system and the challenge is in integrating them as much as possible. The dilemma results from the fact that integrated indicators should be both as general as possible to allow comparisons across fisheries and areas, and as specific as possible to be of practical use in management. This difficult requirement indicates that substantial research should be devoted to the issue.
Sustainability reference systems can be established to compare policy frameworks or management options. Representation of fisheries on SRSs with indicators of change and estimates of confidence limits could give a dynamic representation of its position and evolution in sustainability terms. SRSs could also allow comparison of the positions and dynamics of two or more fisheries, either superimposed on the same SRS (an approach limited by practical complexity of the design) or represented separately on comparable SRSs. It is important to stress (with Prescott-Allen, 1996) that an SRS "reading is a means to an end and not the end itself", that its purpose is to generate debate and serve as a basis for action. In this paper it has been suggested that the SRS be made dynamic and be included as a tool in the management implementation and evaluation cycle.
The main difficulty resides in scaling and aggregation of the potentially available indicators and the scaling of the related value judgements, an area in which achieving agreement is particularly crucial and where both the best scientific evidence and the precautionary approach should play a central role together with considerations of local cultural preferences.
The Pressure/State/Response (PSR) framework provides a way of focusing on the most pertinent indicators even though their potential number remains relatively high. The lack of understanding or complexity of the detailed relationships between indicators (e.g., employment and spawning biomass) and of their quantitative contribution to global sustainability makes the integration a perilous exercise. Hopefully, what is relevant in the SRS is not so much the position of the fishery on the reference as its trajectory with time.
In this respect, a particular difficulty is to be faced regarding the response-time of the system (in terms of "state") to changes in pressure: the amount of inertia inherent in the system should not be underestimated. A representation of transient situations is needed and the indicators of change proposed in this paper could integrate such situations - for example, by representing the present catch per unit of effort as well as the direction and speed at which it is changing due to a management response taken some time ago. The issue is compounded by the fact that biological and economic or social response times are basically different. As a consequence, a management measure or set of measures will have different effects on the various indicators, at different rates, creating potential distortions in the representations.
Much more work is therefore required to test the simplistic representation given by the SRS on particularly well documented fisheries before accepting it generally. An important work is also necessary, at international level, to codify the indicators and the related methodologies and scales before they can become of general use, particularly as a basis for eco-labelling or certification.
ACKNOWLEDGEMENTS
This paper has benefitted from very useful comments of my colleagues Jeff Tschirley and Jorge Csirke to whom I express my most grateful thanks.
REFERENCES
Caddy, J.F. and Garcia, S.M. 1982. Production modelling without long data series. FAO Fisheries Technical Report 278, Suppl. FAO, Rome. pp. 309-313
Caddy, J.F. and Mahon, R. 1995. Reference points for fisheries management. FAO Fisheries Technical Paper 347. FAO, Rome. 83 p.
Calamari, D. and Naeve, H. 1994. Review of pollution in the African aquatic environment. CIFA Technical Paper 25. 118 p.
Dahl, A.L. 1996. Measuring the unmeasurable. Our Planet 8 (1): 29-33. UNEP, Nairobi.
FAO. 1995. Precautionary approach to fisheries. Part 1. Guidelines on the precautionary approach to capture fisheries and species introductions. FAO Fisheries Technical Papers 350/1. FAO, Rome. 52 p.
FAO. 1996. Precautionary approach to fisheries. Part 2. Scientific papers. FAO Fisheries Technical Paper 350/2. FAO, Rome. 210 p.
Garcia, S.M. 1984. Exemple d'application des modèles composites a la Méditerranée espagnole. FAO Rapports sur les Pêches 305. FAO, Rome. pp. 97-105
Garcia, S.M. 1994. The precautionary approach to fisheries with reference to straddling fish stocks and highly migratory fish stocks. FAO Fisheries Circular No. 871. FAO, Rome. 76 p.
Garcia, S.M. 1996. The precautionary approach to fisheries and its implications for fishery research, technology and management: An updated review. In: FAO (1995): Precautionary approach to fisheries. Part 2. Scientific papers. FAO Fisheries Technical Papers 350/2. FAO, Rome. pp. 1-76.
Grainger, R.J.R. and Garcia, S.M. 1996. Chronicles of Marine fishery landings (1950-1994). Trend analysis and fisheries potential. FAO Fisheries Technical Paper 359. FAO, Rome. 51 p.
Gulland, J.A. and Boerema, L.K. 1973. Scientific advice on catch levels. Fishery Bulletin 71(2). FAO, Rome. pp. 325-335
Hilborn, R. and Walters, C.J. 1992. Quantitative Fisheries Stock Assessment. Routledge, Chapman and Hall. 570 p.
Pontecorvo, G. and Schranck, W.E. 1995. Commercial Fisheries: the Results of Stochastic Supply and Economic Uncertainty. A paper presented at the Columbia Resources Seminar on Sustainable Development and a Managed Resource: the Current Crisis in Commercial Fisheries. Arden House. 5-6 May 1995. British Columbia University, Morrilton, USA. 61p.
Prescott-Allen, R. 1996. Barometer of Sustainability. What it's for and how to use it? The World Conservation Union (IUCN), Gland, Switzerland. 25 p.
Shuh, E.G. and Archibald, S. 1996. A framework for the integration of environmental and sustainable development issues into agricultural planning and policy analysis in developing countries. In: Integration of Sustainable Agriculture and Rural Development Issues in Agricultural Policy. S.A. Breth (ed). Winrock International, Vancouver, Canada. pp. 3-44.
WCED. 1987. Our Common Future. World Conference on Environment and Development. Oxford University Press, Oxford, UK. 400 p.
APPENDIX 1
DEFINITIONS
The following definitions apply to this paper and do not pretend to be universally agreed:
Criterion: Condition to achieve some development objective, defined through critical review of scientific information 1 (modified from Calamari and Naeve, 1994). A set of criteria would provide a system of reference within which to assess or judge the state of the exploited system as reflected by indicators.1 In fisheries, and according to the Law of the Sea, criteria would have to be based on "the best scientific information available".Indicator: A variable, a pointer, an index of a complex phenomenon. Its fluctuations reveal the variations in components of the ecosystem, the resource or the sector. The position and trend of the indicator in relation to the criteria indicate the present state and dynamics of the system. Ideally, composite indicators are needed, the position and trajectory of which within a system of reference of related criteria would allow simple holistic assessment of sustainability.
Reference Point: A reference point indicates a particular state of a fisheries indicator corresponding to a situation considered as desirable (Target Reference Point, TRP), or undesirable and requiring immediate action (Limit Reference Point, LRP, and Threshold Reference Point, ThRP).
Standard: criterion which has been formally established and enforced by an authority.
APPENDIX 2
TYPES OF INDICATORS
Fishery-related indicators
Yield-related indicators
Criteria:
Maximum Sustainable Yield (MSY); Maximum Constant Yield (MCY), Long-Term Average Yield (LTAY);Indicators:
Catches, Catch value, Pelagic/Demersal ratio (P/D), Ratio Yield/MSY, MCY, LTAY, etc.
Capacity-related indicators
Criteria
FMSY, FMCY, FLTAY, FMEY, F0.1, FOY (undetermined)Indicators
Fishing effort (f), Fishing intensity (f/unit area), Fishing mortality (F). Ratios between current and target f (or F): f/fMSY, f/fLTAY, etc. Changes in the ratio between searching time and effective fishing time could be very symptomatic of changes in abundance. Progressive shifts in fishing areas could indicate resource modification.
Other economic indicators
Criteria:
Economic self-sustainability; Conservation of long-term aggregated welfare potential; Maximum value (?); Maximum rent, Maximum aggregated profit, Subsidies = 0Indicators
Investment; natural resource valuation scale available; total natural asset value; level of subsidies; ratio of subsidies to capital value; ratio of loans to investment; ratio of present to maximum catch value (?); ratio of present to maximum profit (?); ratio of extracted to maximum or expected rent; poverty, wealth, average age.
Technological indicators
Criteria
Environmentally-friendly gear; highest possible selectivity; lowest possible discard rate or zero-discard option (forced landing); best available and locally affordable technology. Lowest ozone-depleting gas emissions.Indicators
Existence of Prior Consent (PC) or prior Authorization (PA) procedures. Lists of acceptable gear. Gear regulations addressing selectivity, by-catch of juveniles, and discards (Grid systems, protected nursery areas).
Social indicators
Criteria
Maximum sustainable employment (MSE?); minimum social unrest; equitable (comparable) revenues; acceptable allocation of wealth; safety on board; food security.Indicators
Coastal populations; employment rate, sectoral emigration and immigration; age; frequency and violence of conflicts, non-compliance index; ratio between fisheries and other revenues; rate of boat and life loss (% loss per year); availability of fish/caput; revenues.
Institutional indicators
Criteria:
Research and decision-support capacity. Specific legislation; Effective peoples' participation. Effective monitoring, control and surveillance; establishment of management committees, allocation of wealth and dispute resolution mechanisms.Indicators
Research staff and budget. Information systems (e.g., GIS, databases). Contribution of research to decision making and to assessment of management performance. % of fisheries covered by management committees; degree of participation (in information collection, option analysis, decision making, enforcement). Number and role of NGOs. Duration and severity of conflicts (?).
Ecosystem-related indicators
Resource biomass criteria:
Virgin biomass (BV), Minimum biological limit (MBAL), 0.3BV, BMSY, BMCY, etc.Indicators
Biomass of target and non-target species. Ratios of key variables to target or limit reference point (B/Bmsy, B/Bv, etc.). Indicators could be fishery-related measures of biomass such as catch per unit of effort (cpue) or independently obtained (such as through trawl and acoustic scientific surveys). Changes in distribution area could be used as a proxy to changes in biomass for small pelagic and some other resources.
Resource demographic structure
Criteria
Lc50, Lm50,, LF=0, tc50, tm50,
.
Indicators
Length or age composition and average length or age; ratio of average length or age to length or age at first maturity (Lm50 and tm50). Ratio of present average length () to pristine average length or age (LF=0,) eventually simulated with F = 0. Sex-ratio where relevant (e.g. marine mammals). School size where relevant (e.g., small pelagic resources). Fat index (e.g., in small pelagic fish used for reduction).
Biological diversity
Criteria
Minimum possible species loss. Minimum loss of genetic diversity.Indicators
Existence of Protected Marine Areas (PMA). % littoral area protected, totally, partially.
Existence of germplasm conservation scheme. Biodiversity index.
Water quality indicators:
Criteria:
"Pristine" conditions or conditions at an agreed reference point in time. Conventional criteria, as established by international environmental conventions, Codex Alimentarius, etc.Indicators
Transparency (Secchi values); satellite colour scanner indications; Algae index; release of nitrogen components and phosphorates. Other global pollution indicators. Population density (average and seasonal peaks). Absolute values or ratios.
Critical habitats indicators
Criteria:
Seagrass beds, Mangroves, mudflats, coral reefs, "Pristine" habitat area (or, area at an agreed reference point in time), state.Indicators
Ratio between residual area and "Pristine" or reference area, area of live and dead coral, grass density, species diversity indexes, other indexes of condition.
APPENDIX 3
CATCHES AS AN EXAMPLE OF A SUSTAINABILITY INDICATOR
Type catches: are a response indicator. They relate to the time-related development of a fishery from an undeveloped to a fully developed or overexploited state. Catches describe the main output from a fishery rather than the ability of the resource to provide that output in a sustainable way. As such, catches are very crude indicators of "sustainable use" as they take no account of inputs such as fishing effort and recruitment variation.
Purpose: To monitor the main output of the activity of the fishery sector. In practice, catch data alone are likely to provide only a very crude and unreliable indication of sustainability and so they should be utilized with other information where possible (e.g. on effort, mortality or biomass). However, long data series could be perfectly used to establish retrospective diagnosis of the evolution of fisheries and, by inference, of the present state (see Grainger and Garcia, 1996, for example).
Policy relevance: If catches have been maintained at a fairly steady level for many years during which fishing effort has not increased, it is likely the fishery is sustainable unless environmental conditions change. In the absence of better information, the catch level in such a situation is sometimes used to set a total allowable catch in order to prevent a diversion of fishing effort into the fishery. Catch data for a short time period should never be used as indicators as unusually high recruitment may support a level of fishing in the short term which would not be sustainable in the long term.
Relevance to sustainable development: Irrespective of what other indices are available, catch trends should always be monitored. Declining catches which cannot be explained by factors other than exploitation (e.g., by reduced fishing activity) should be taken as a warning of a possibly unsustainable level of exploitation, and assumed to be unsustainable until such time as more reliable information would become available to indicate otherwise.
Linkages between indicators: In theory, catch per unit effort is an index of stock abundance which can be used to monitor the response of a stock to the development of a fishery. In practice, however, the determination of effective fishing effort is difficult and the relationship between fishing effort and fishing mortality changes as fishing technology develops, with the result that catches alone can sometimes reflect stock abundance as well as catch per unit effort. Combined with other data or indicators, catches can provide useful indexes: for example, the ratio between present and historical maximum, or the ratio between landings of demersal and pelagic species (Grainger and Garcia, 1996).
Targets: Catch limitations are commonly used to control fishing mortality and total allowable catches may be considered targets by industry.
International conventions and agreements: No international conventions propose the use of catches as indicators of "sustainable use" but FAO is mandated to collect fishery statistics from its member countries and to provide summaries. The Agreement for the Implementation of the Provisions of the United Nations Convention on the Law of the Sea of 10 December 1982 Relating to the Conservation and Management of Straddling Fish Stocks and Highly Migratory Fish Stocks, as well as the Code of Conduct for Responsible Fisheries, state that stock-specific reference points should be applied for this purpose, but that catch and other relevant data on the fisheries should be collected and used to assess the impact of fishing on the resource.
Data requirements: Statistics on landings by fishery and by stock. The catch represents the total removal from a fish stock due to fishing activity. Total catch includes the retained catch which is consumed on board or subsequently landed, the discarded catch as well as incidental deaths resulting from the fishing activity. In practice, however, catch data usually only include the landed component because of data shortcomings.
Methodology: There are no internationally agreed methodologies. Catch data have traditionally been collected through census approaches and are still collected in this way (from vessels records of landings or auction hall records, particularly when catch quotas are used for management). They are increasingly calculated from catch assessment surveys and stratified sampling techniques yielding statistical estimates with confidence limits.
Availability: Catch data, albeit with varying degrees of bias and precision, are usually the most widely available data, at both national, regional and international levels.
APPENDIX 4
MAXIMUM SUSTAINABLE YIELD AS A SUSTAINABILITY CRITERION
Type: Maximum sustainable yield (MSY) is a state criterion and an internationally agreed management reference point, i.e. a point by reference to which the state of the fishery is assessed.
Purpose: Maximum sustainable yield is enshrined in UNCLOS and although no longer accepted as a valid (and precautionary) development target, remains a very important benchmark for management and a minimum target for depleted resources rehabilitation (in the 1995 UN Agreement and in the 1995 FAO Code of Conduct). There are numerous alternative benchmarks at lower rates of fishing (c.f. Caddy and Mahon, 1995; Garcia, 1996) but these are yet to be formally agreed and codified, so MSY remains a necessary universal benchmark. Used in relation with an appropriate indicator of level (such as yield, or effort) it indicates where the fishery system stands in biological sustainability terms.
Relevance to policy: Being enshrined in UNCLOS, MSY is a necessary reference for fisheries policy. The state of the system in relation to the state corresponding at MSY will be one of the key factors in assessing the sustainable nature of the policy. All international fishery agreements, whether legally binding or voluntary require a reference to MSY. MSY is also recommended as a minimum target for resource rehabilitation policies.
Relevance to sustainable development: Although fisheries are sustainable at various levels of development, the risk of collapse (with its related socio-economic dislocations) increases as the exploitation rates (and stress) approach the MSY level and increase beyond it. If a resource biomass is at or below that corresponding to MSY, or if the fishing effort or fishing mortality is at or above that corresponding to MSY, there must be serious concern that the resource may be severely affected and is likely to be overexploited. MSY conditions imply a level of fishing effort in excess of economically optimal harvesting. In addition MSY conditions usually correspond to increased resource variability and uncertainty for the manager and industry, increasing risks and potential costs of error.
Linkages with other criteria and indicators: Determining MSY requires that a time series of catches (another potential indicator) is available. Combined with other indicators, MSY may be used as indicator of level. Where MSY has previously been determined, with its corresponding level of effort (FMSY) and biomass (BMSY), it is possible to monitor the fluctuations of F/FMSY, B/BMSY and Y/YMSY with time as a measure of performance. Considering that BMSY is usually assumed to be about half of the virgin biomass (Bv), the ratio B/Bv could provide a closely related indicator when indices are available for both magnitudes e.g. using present catch rates as proxy for B and catch rates at very low fishing pressure at the beginning of the fishery as proxy for Bv. When B/Bv = 0.5 the fishery has reached approximately the MSY level. Although MSY is determined solely on the basis of yield and effort, its position in relation to the maximum economic yield (MEY) is known and limited inferences can be made even though the economic data may not be available.
Target: The MSY concept is a macro-level indicator, irrelevant to individual industry operators but very relevant for governments (in complying with their duty of care) and for fishery management organizations which can use it to develop limit reference points (LRPs) for management (i.e. upper limits to the rate of fishing (or lower limits to the spawning biomass)). It is specified in relevant international conventions that when LRPs are approached, action should be taken to ensure they are not exceeded.
Related international conventions and agreements: As mentioned before, MSY is referred to in all international conventions related to fisheries, including UNCLOS, the 1995 UN Agreement and the 1995 FAO Code of Conduct (Article 6).
Data requirements: Unbiased catch data, corrected for discards, as well as stable indices of fishing pressure or fishing mortality. The effort level should be given in standard units adjusted for changes in fleet fishing power over time or changes in geographical extension of the fishery. If reliable time series are not available, contemporary catch and effort data for a number of comparable fisheries with the geographical extension of the stock could be used, assuming equal productivity.
Appropriate methodology: A number of methodologies are available and have been used such as more or less sophisticated surplus production models elaborated by well known scientists such as Schaefer, Gulland, Fox, Pella and Tomlinson (Hilborn and Walters, 1992). The use of composite space-structured models is illustrated in Caddy and Garcia (1982) or Garcia (1984).
Availability: MSY information is available for a large number of resources worldwide. The information available is of very uneven quality and has not been estimated with a uniform methodology. It is rarely given with confidence intervals. The information is often outdated and, with the changes in the ecosystem provoked by fishing (e.g., reduction of predators' abundance), it is sometimes doubtful that available information is still valid. In addition, in areas of high natural variability (e.g., upwelling areas), a long-term average value for MSY is usually not available and would, in any case, be of little use for management.
