This manual describes how to manage sampling programmes for routine data collection. Routine data are collected in a never-ending programme and are used to monitor fisheries. Although the programme is continuous, it does not mean the programme has a fixed design. It should be able to adapt to changes in technology and activities, as required. Setting up and maintaining a sample-based routine data collection programme is the subject of this manual.
Although related to the Guidelines for the routine collection of capture fishery data (FAO, 1999), this manual contains different information. The Guidelines recommend general methodologies, but do not give instructions on how they might be implemented. There is no universal approach to the implementation of fisheries data collection programmes. This manual gives examples of how parts of the methodology covered by the guidelines might be implemented in practice. The present manual is not comprehensive in the sense that it addresses all topics covered by the guidelines. Rather, it attempts to give representative examples that illustrate the basic principles. Therefore, the specific methodologies can rarely be applied directly, but it should be possible to modify them to match any particular case.
A prerequisite for a successful design of a data collection programme is that the designer understands the objectives of the data collection. It is not always an easy task to link the methodology of data collection to the objectives, as there are usually many objectives. The present manual cannot present the entire theory on the use of fisheries data (fish stock assessment, country sector profiles etc.), but tries to introduce just enough theory to justify the data collection and to define the data requirements.
Design of a data collection programme involves solving a variety of problems through combining the techniques covered in this manual with practical sense, historical knowledge of the fishing sector and understanding of political issues. The present manual is based on a number of assumptions about the structure of the fishery to which the methodology is applied. Without the knowledge about the assumed underlying fisheries structure, it may be difficult for some readers to understand the particular design of the sampling programme introduced here. The author does not consider it possible to construct a “generic data collection methodology” which would apply to any nation or fishery of the world. The author (although usually associated with Denmark and ICES) has primarily used experience from East Asia, in particular from Viet Nam, to write this manual. The author, being a mathematician by education, has experience in fish stock assessment, fisheries management and bio-economics of fisheries, and the methodology suggested is (naturally) influenced by this basis of experience.
During the process of setting up a nation-wide data collection system for the commercial marine capture fisheries of Viet Nam, national collaborators, the author and consultants completed many tasks involving problems of various complexities. The tasks covered a wide range of disciplines. This manual attempts to cover all the major technical tasks faced in Viet Nam, with the main emphasis on the tasks considered problematic or which resulted in lengthy discussions within the group of programme designers. The identification of species and the recording of biological parameters (such as “body length”) did not create major problems in Viet Nam, but may require a training programme in other countries. These methods are outside the scope of manual, but may be found in other texts.
The history of marine fisheries is full of incidences of overfishing and stock collapse resulting in closures of fishing industries and bankruptcies. Therefore, it may be argued that the primary objective of fisheries management and planning of fisheries development is to avoid over-investment. Often, investment in the fisheries sector was based on very limited knowledge about the ecosystem, the fish stocks, the fisheries sector and the fisher communities. This lack of knowledge made investors and managers behave optimistically or irresponsibly. In such cases of limited background knowledge, a “Precautionary Approach” should replace the “optimistic” approach. Knowledge and understanding about the real underlying ecosystem, and underlying dynamics of fish stocks are always limited to the system perceived by the researchers and managers.
The precautionary approach, sustainable development, rational exploitation and responsible fishing have been given a central place in international conferences and agreements devoted to the environment of fisheries. Some of the more relevant definitions and statements were given in the “Code of Conduct for Responsible Fisheries”, CCRF (FAO, 1995), and in other FAO publications.
International agreements show that a consensus view exists that a “precautionary approach” is required for the management of fisheries. However, there may be controversial interpretations of what the precautionary approach actually is. Paragraph 7.5.1. of the CCRF is devoted to the precautionary approach. It stipulates:
States should apply the precautionary approach widely to conservation, management and exploitation of living aquatic resources in order to protect them and preserve the aquatic environment. The absence of adequate scientific information should not be used as a reason for postponing or failing to take conservation and management measures (The concept of “reference points” is discussed in Section 2.3).
and in respect to paragraph 12.13 of Article 12, on fisheries research:
"States should promote the use of research results as a basis for the setting of management objectives, reference points and performance criteria, as well as for ensuring adequate linkage between applied research and fisheries management.
A prerequisite for research is data collection. The backbone of data collection for fisheries research is routinely collected data. In its Article 5(c), the UN Agreement on the Conservation and Management of Straddling Fish Stocks and Highly Migratory Fish Stocks stipulates that the precautionary approach should be applied in accordance with Article 6. One paragraph is of particular relevance in the context of data collection:
6.3. In implementing the precautionary approach, states shall:
a) Improve decision-making for fishery resource conservation and management by obtaining and sharing the best scientific information available and implementing improved techniques for dealing with risk and uncertainty;
....
d) Develop data collection programmes to assess the impact of fishing on non-target and associated or dependent species and their environment, and adopt plans, which are necessary to ensure the conservation of such species and to protect habitats of special concern.
The adoption of the precautionary approach has considerable implications for fishery management agencies and the fishing industry. Scientific advice to fisheries managers should allow for uncertainty in both the understanding of the state of the stocks and the effects of future management actions. This implies that when less is known, fishery management agencies should be more cautious. This requires a management approach less influenced by short-term considerations, and more concerned with long-term sustainability.
Socio-economic factors to be considered in establishing objectives for the management of fisheries might, for example, imply the sustainable maximisation of yield, or of employment, either in the fishing industry or in the more general fishing sector. Unfortunately, all desirable objectives cannot usually be met simultaneously, and one of the main roles of fishery management agencies in a precautionary approach would be to derive trade-offs between competing objectives in consultation with interested parties. Fishery management agencies could, for example, pursue economic goals such as high profitability (which implies low exploitation rates and high fishing efficiency) social goals such as high employment (which require higher exploitation rates and/or lower efficiency) or some quantified trade-off between these conflicting objectives. Whichever approach is taken, it will be necessary to quantify objectives and trade-offs it they are to be translated into measurable factors such as levels of fishing mortalities. It can be concluded that international standards consider data collection a prerequisite for responsible management and development of fisheries. It is irresponsible to initiate any major development of fisheries before a solid basis of knowledge has been made available. The more limited the database, the more strictly the “precautionary approach” should be observed.
Data collection in fisheries is a rather wide field, which may deal with the theory of statistical analysis, design of relational databases, measurements of biological parameters, estimation of biological, technical or economic parameters, analyses of fisheries household survey etc. It is not the intention that the present manual should cover all fields. Other textbooks and manuals in the scientific literature already cover many of these topics in detail. Therefore, topics that are well covered by standard textbooks will be only briefly mentioned here.
The manual is intended to assist in setting up a data collection system mainly from the practical and organisational point of view. This includes the allocation of limited resources (labour and funds) to tasks in the programme, as well as development of the human resources.
The methodology of data collection is dealt with at a level, which does not assume background knowledge in any scientific discipline.
The main questions addressed in the manual are which fisheries data to collect, where and when to collect them.
Only data collected from commercial marine capture fisheries are considered (data from fresh water fisheries, cultured fish and experimental fishery are excluded). The methodologies are most appropriate for a tropical, developing country, with many small (artisanal) vessels and few large (industrial) vessels.
The methodology is the “sample-based approach” - the manual does not deal with a methodology which assumes complete enumeration (such as the systems based on compulsory filling in of logbooks / sale-slips applied in many industrialised and some developing countries). The present data collection methodology attempts to utilise whatever information can be obtained in practice in a developing country, including skippers and buyer's notebooks.
Processing and storing of data (fisheries databases) is partly covered because they are considered tools to control, evaluate and improve the data collection programme. The design of a fisheries database is considered an integral part of the design of a fisheries data collection programme, and the two tasks should not be separated. In order to analyse the collected data (for example, evaluating their quality), some knowledge of databases is required. All staff of a data collection programme should have some knowledge of relational databases, and some staff must be experts. However, a comprehensive introduction to the theory on design of relational databases is outside the scope of the manual. Readers interested in further studies on relational databases are referred to the relevant literature (Chapter 8).
The following topics, which might have been included (and which can indeed be found in some manuals of fisheries data collection) are not covered by the present manual. That the topics are not covered does not imply that they are considered to be less important, but they are considered covered adequately elsewhere. These include:
The practical aspects on how to measure technical parameters. For example, how to identify species, how to measure length, weight, maturity stage, age of individual animals, dimensions of fishing vessels, fishing equipment and fishing gears.
Economic data such as prices of landings, costs of fishing, investments etc.
Specialised environmental data. Only data directly related to the fishing operation are included in the programme.
Data from processing, marketing, distribution, quality control and import/export sectors. Some data from these sources may constitute valuable supplements to the data collected from the harvesting sector, and some brief remarks on the use of data from processing plants are given in Section 5.4.
Data from experimental fishing or from research vessel surveys. There are already several manuals, including some published by FAO on collection and analysis of data from experimental fisheries and research vessel surveys. Data obtained from experiments or surveys can often be combined with data from a commercial fishery in, for example, fish stock assessment.
The mathematical theory on sampling. For the theory of statistics (calculation of confidence limits etc.) and sampling techniques, the reader is referred to standard textbooks (see Chapter 8). Only a brief (non-mathematical) introduction of the “Neyman criteria” is given (see Section 4.3.1).
The manual follows the same structure as the FAO (1999) Guidelines for the routine collection of capture fishery data (FAO Fisheries Technical Paper 382; Figure 1.2.1). However, this manual concentrates on practical aspects, rather than the general framework for developing a programme.
Chapter 2 gives a brief discussion of the objectives of data collection in general terms. It also covers who has the responsibility for the fisheries data collection and the database. The allocation of responsibility is a prerequisite for the implementation of a programme and the responsibility is often associated with the use of the data. The idea of management units and reference points as management indicators is also introduced.
Chapter 3 introduces the basic types of data that are collected. It presents some important concepts of fisheries data collection methodology and gives some of the theory behind the use of the data, such as fish stock assessment and bio-economics. The methods used to extract information from the data are relevant to the way the data are collected.
Chapter 4 is concerned with the design of sample-based data collection programmes. It discusses the different types of stratification of the fisheries sector that optimise data collection, as well as stratification required to meet requests from clients. Often the aim of the programme is to obtain as accurate as possible statistics such as total catches from partial coverage. This chapter also explains how samples are raised to these totals.
Chapter 5 presents some data collection methods, mainly through considering what and where data are collected. Suggested data collection forms to be used by enumerators are presented
Chapter 6 gives the basic considerations and design of a fisheries database. In addition, Section 6.7 deals with validation of data, and Section 6.8 with the output from the database (reports). Section 6.9 describes the “Fisheries sector profiles”, which contain background information about the fishing sector. This overall description of the sector is the minimum information required by fisheries managers and administrators.
Figure 1.2.1 Setting up a data collection programme follows from identifying data needs through to working out how the data should be collected. In designing the programme, all options should be carefully considered. (From FAO, 1999).
Chapter 7 discusses various aspects of the implementation of sample-based data collection programmes from scratch. Development of human resources is an important element, and this chapter concentrates mainly on this issue. Also, the budget and the documentation of the methodology are discussed.
The concepts are illustrated extensively with examples. Chapter 3.4 introduces a hypothetical, but representative, tropical developing country, which is used to provide an fisheries structure for many of the subsequent examples. Chapter 6.10 presents a minimal fisheries relational database for demonstration of basic principles of fisheries databases. The Fisheries Demonstration DataBase (FDDB) is constructed to illustrate the full-scale fisheries database, “VIETFISHBase” developed by the author and collaborators for the Viet Nam commercial marine fishery and experimental fishery during 1997 – 1998. The FDDB is a combination of various fisheries databases developed by FAO projects.
Finally, Chapter 8 contains a list of literature suggested for further studies.
Although the structure of this manual suggests a linear process from design to implementation, in reality this is only partially true. There are two important aspects, which designers of a data collection programme will need to bear in mind:
Implementing a data collection programme is an iterative process. So the implementation should be regularly reviewed and modified. Each completed year of the programme should add to the accumulated experience, which is used for planning and implementation in the following year. We shall come back to the iterative process in Section 7.6.
Often it makes sense to design different parts of the programme concurrently rather than in sequence. For example, the database system should be developed alongside the data collection methods. The form design will often be integrated with design of the database. The database will also be used for checking and validating input, and direct input by fieldworkers may eliminate the need for paper forms at all.
The manual is intended for professionals in fisheries research, administration or management. The reader is assumed to be familiar with fisheries. Otherwise, no assumptions are made regarding the reader's background.
The primary target groups are:
The secondary target groups are:
Individual workers in the primary group will usually also belong to one of the secondary target groups, whereas the secondary group may encompass a much larger group of individuals than the primary group. It has been attempted to introduce each field of expertise so that workers in other fields understand key issues of the data collection programme.
