Standardization of Fishing Effort in Qatar Fisheries: Methodology and Case Studies

Stamatopoulos C^1* and Abdallah M²

¹Consultant of Fisheries Statistics, Department of Fisheries, Ministry of Environment, Doha, State of Qatar

²Consultant of Fisheries Resources, Department of Fisheries, Ministry of Environment, Doha, State of Qatar

*Corresponding Author:

Stamatopoulos C
Department of Fisheries
Ministry of Environment
State of Qatar
Tel: +003-9069360376
E-mail: cstamat@gmail.com

Received date: September 25, 2015; Accepted date: October 21, 2015; Published date: October 27, 2015

Citation: Stamatopoulos C, Abdallah M (2015) Standardization of Fishing Effort in Qatar Fisheries: Methodology and Case Studies. J Marine Sci Res Dev 5:170. doi:10.4172/2155-9910.1000170

Copyright: © 2015 Stamatopoulos C, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Visit for more related articles at Journal of Marine Science: Research & Development

Abstract

Regular statistical monitoring of fishing activities is a prerequisite for effective fisheries management. In the case of artisanal fisheries such a monitoring is often exercised by means of sample-based fisheries surveys in which catch and fishing effort (along with other basic variables) are estimated on the basis of samples relating to landings and boat-gear activity. In most cases the fishing fleet is heterogeneous and hence partitioned into boat-gear categories in each of which fishing units have similar characteristics and performance. Under this scheme catch/effort estimates are computed for each boat-gear category separately and independently of each other. It can then be assumed that in each boat-gear category fishing mortality is proportional to the total fishing effort exerted by all of its fishing units operating together. When it comes to measure the combined effect of the fishing operations of the entire fleet to the exploitation of a fish stock, it becomes apparent that adding together effort exerted by different boat-gear categories is not always meaningful without first applying effort adjustment to increase its comparability. There are various techniques for addressing such situations, the commonest of which is known as “standardization of fishing effort”. In Qatar the National Fisheries Information System (NFIS) has recently incorporated effort standardization routines that combine elements of the normalized relative effort (used by the North Sea Round Fish Working Group, ICES, 1980) with those of relative fishing power developed by Robson (1966). The document presents the methodology in use by NFIS for effort standardization as well as case studies using commercial catch/effort data directly obtained from NFIS. It is envisaged that the selected approach will be further refined in order to increase the role of catch/effort data in research and stock assessment applications.

Keywords

Fisheries statistical monitoring; Sample-based surveys; Catch/effort assessment; Standardization of fishing effort

Introduction

In Qatar the fisheries resources are exploited by artisanal fishing units comprising two fishing vessel types: launches (large boats) and speedboats (or tarads). The launch is a decked vessel usually constructed of wood or fiberglass and powered by an in-board engine; the average trip duration is between 3 and 5 days. The speedboat is an open dory usually of fiberglass construction powered by one or two outboard engines. Due to its smaller size the trip duration is usually one day. It also alternates its fishing gear depending on the species sought; this however is known not to occur during the same fishing trip. All fishing units operate from the four ports of Al Shamal, Al Khor, Doha and Al Wakra. There are about 500 launches and 1000 licensed speedboats. Not all fishing units are active during a month; their operational state is variable and the number of active fishing units (a boat is considered active if it has made at least one fishing trip during the month) is enumerated on a monthly basis.

The top 11 species contributing to Qatar fish landings in 2014 were reported by the National Fisheries Information System (NFIS) to be: Spangled emperor Lethrinus nebulosus (16.2% of total landings) (Figure 1), Narrow-barred Spanish mackerel Scomberomorus commerson (10.5%) (Figure 2), White-spot spinefoot Siganus canaliculatus (8.3%), Pink ear emperor Lethrinus lentjan (6.6%), the Orange spotted grouper Epinephelus coioides (6.0%), Haffara Seabream Rhabdosargus haffara (4.3%), King soldier bream Argyrops spinifer (3.7%), Painted sweetlips Diagramma pictum (3.5%), Gold toothless trevally Gnathanodon speciosus (3.4%), Orangespotted trevally Carangoides bajad (3.1%) and the Eastern little tuna Euthynnus affinis (2.2%).

Figure 1: Spangled emperor (Lethrinus nebulosus).

Monthly and annual landings of statistically monitored species, 57 in all, are systematically reported to and reviewed by the Fisheries Department of the Ministry of Environment, along with other relevant data that are standard part of the statistical database in use.

Until mid-2012 statistical monitoring was limited to catch information collected at the Central Auction Market in Doha. Due to its limitations in data scope and coverage a new system, the internet supported National Fisheries Information System (NFIS), was implemented in 2012. NFIS is the first of the four principal components of the Qatar Government research project “Sustainable Management of Fisheries Resources” which is being executed by The Ministry of Environment, Department of Fisheries and in close collaboration with the Qatar University Environmental Study Centre, the Qatar Science and Technology Park and The Prince’s Charities’ International Sustainability Unit of the UK, which provides technical advice whenever it is required. The project’s main components comprise: (i) Implementation of a web-based National Fisheries Information System; (ii) Establishment of a Marine Spatial Planning System; (iii) Development and implementation of a fishery management plan based on Maximum Sustainable Yield and, (iv) Development and implementation of a Communication Plan. The web-based NFIS has been regularly operating since September 2012 for the systematic recording of sample data on catch, fishing effort, catch per unit effort, fish size and prices; all such data are collected for each boat-gear category and at all four fishing ports. The data are subsequently fed into an online database accessible by user groups for statistical analyses and reporting. At present NFIS is operating at full capacity and contains data of acceptable accuracy (the threshold of accuracy is 90%) covering the period September 2012 to date. It offers a wide variety of online reports that are supplemented by easy-to-use plotting and tabulating utilities.

Owing to the heterogeneity of the artisanal fishing fleet NFIS has partitioned it into four categories of fishing units of similar characteristics and performance and in a manner that catch/effort estimates are computed separately for each boat-gear category and independently of each other. The four boat-gear categories of NFIS comprise:

1. Launches using traps;

2. Launches using kingfish nets;

3. Launches with miscellaneous gear and;

4. Speedboats (tarads) with miscellaneous gear.

It is generally accepted that when working with a specific boat gear category (for instance launches with traps) fishing mortality is proportional to the total fishing effort exerted by its fishing units. When it comes to measure the combined effect of fishing operations of the entire fleet to the exploitation of a fish stock, it becomes apparent that adding together effort exerted by different boat-gear categories is not always meaningful without first applying effort adjustment to increase its compatibility. There are various techniques for addressing such situations, the commonest of which is known as “standardization of fishing effort”. Maunder [1] gives a more general description of effort standardization as the “the ability to use catch rate data as an index of abundance by removing the impact on catch rates of changes over time of factors other than abundance”.

In Qatar the National Fisheries Information System has recently incorporated effort standardization routines that combine elements of the simple (if not very recent) normalized effort (used by the North Sea Round Fish Working Group, ICES, 1980) and relative fishing power developed by Robson [2].

To be sure the existing literature offers a plethora of other more advanced methods for the standardization of catch and effort data which involve fitting statistical models to the catch and effort data. The first examples of these methods were by Gavaris [3] and Kimura in which General Linear Models (GLM) were used. Moreover the last two decades have seen a proliferation of new methods to standardizing catch and effort data, most of which extend these methods to various degrees. For instance Generalized Additive Models were used by Bigelow et al. and Rodriguez-Marin et al. Generalized Linear Mixed Models extend the GLM approach by allowing some of the parameters in the linear predictor to be treated as random variables. Several analyses of catch and effort data [4] made use of GLMM techniques.

The choice among these methods (an excellent review of which is made by Maunder [2] is based on an evaluation of the underlying assumptions of the models and the type of appropriate statistical tests and diagnostics to be employed. In addition to methodological aspects there are several operational criteria and constraints concerning the type, amount and quality of data to be used [5]. In the case of Qatar it was considered that at the first stages of effort standardization all analyses should be based exclusively on regularly collected catch/effort data from commercial fisheries and that the effort standardization routines should be part of the NFIS report generator [6]. Such being the case the concepts and approaches used by Robson [1] and ICES (1980) seemed to constitute a good and practical basis for developing the presented method.

The need for effort standardization was first pointed out by the Steering Committee of the Sustainable Management of Fisheries Resources project and was followed up by the Fisheries Department of the Ministry of Environment.

Thanks to the collective effort made by field staff and the national experts of the Fisheries Department the presented methodology was repeatedly tested using data of good quality, completeness and accuracy [7,8]. It should also be noted that the present study is only the first step in introducing effort standardization as a regular operational component of NFIS; the approach in use will be further refined when catch/effort data involving more years have been made available [9,10].

The effort standardization approach used by Qatar may be of potential interest to other neighboring countries in the Gulf region which operate similar fleets [11,12]. Effort standardization on a regional basis should not present a major problem if data protocols were setup permitting comparability of nationally available catch/effort data. Such activities would be part of ongoing regional cooperation and considerably facilitate regional catch/effort assessment for important shared stocks.

Materials and Methods

Primary variables of the study

In this study the fishing effort exerted by a fishing unit during a fishing trip is measured by the duration of the trip and referred to as “boat-gear days”. If there are m boat-gear categories and the statistical monitoring system produces 12 monthly catch/effort estimates per boat-gear category (as is the case with NFIS) then over a reference period of n years there will be (m x 12n) monthly effort estimates E_i,j, i=1…m; j=1…12n.

Along with fishing effort the system estimates monthly catch C_i,j and Catch-Per-Unit-Effort CPUE_i,j.

It should be noted here that NFIS treats combined CPUE’s as weighted averages and not as simple arithmetic means of their components. For instance to combine monthly CPUE values of the same boat-gear category into a an annual CPUE, the standard NFIS procedure is to re-calculate the monthly catch and effort values involved according to the standard formula Σ(Catch) /Σ(Effort) .

Table 1 illustrates an example of a full set of NFIS catch/effort estimates for 2014 which involves the three primary variables described above.

Table 1: NFIS catch/effort data for 2014 (all species) – Accuracy of estimates: 91.7%.

Computational steps in effort standardization

The objective of the presented method is to achieve effort compatibility when different boat-gear categories are combined together. Specifically, its two tasks are:

1. Producing total standardized effort of combined boat-gear categories;

2. Computing standardized CPUE’s for combined boat-gear categories;

It should be noted here that the example given in this section and summarized in Tables 1 and 2 treats catch as a whole and without focusing on a specific fish stock; such a consideration is used only temporarily with the sole purpose of facilitating the presentation of the computational steps in effort standardization. In Section 3 that describes the results of the study readers will be presented with two case studies dealing with Spangled emperor (Lethrinus nebulosus) and Narrow-barred Spanish mackerel (Scomberomorus commerson) respectively; these are the two top species of the 2014 landings in Qatar.

Table 2: Standardization of the NFIS catch/effort of Table 1.

The method starts by considering the compatibility of CPUE’s of different boat-gear categories. Since these involve incompatible effort values in the denominator they cannot be combined at monthly or annual levels (notice the absent values for effort and CPUE in the totals line in Table 1). This happens since they are viewed as weighted averages over a period of a month or a year.

On the other hand each of these CPUE’s could be temporarily viewed as the representative catch by just one boat from each boat-gear category during one day.

Using this second concept for monthly CPUE’s by boat-gear category and over 12n periods, a 2-dimensional array of daily yields P_i,j can be formed where:

i=1…m (boat-gear categories);

j=1…12n (monthly estimates).

To be noted that the notation has changed from CPUE to P since a CPUE is expressed in Kg / boat-gear day while the newly assumed daily yields P are in Kg.

The method proceeds with the following notations and computations:

The sum of all daily yields is given by:

(1)

The arithmetic mean of all daily yields is given by:

(2)

Working with a boat-gear i it is found that its total daily yield is:

(3)

and the arithmetic mean is:

(4)

The overall arithmetic mean shown in (2) is now assumed to represent the overall daily yield of a new (and hypothetical) boat-gear category. To compare the overall performance of each actual boat-gear to the new hypothetical one the following ratio is used:

(5)

where and are obtained from (4) and (2) respectively.

In the presented study this ratio is referred to as standardization factor since it is used for converting actual effort into a standardized one. Once calculated, the standardization factor f_i is considered to remain the same across all periods. Consequently each effort cell E_i,j representing effort of boat-gear i in period j can be converted to standardized effort using the expression:

(6)

Adding up all m standardized (thus addable) monthly effort values for a period j will result in a monthly standardized effort E_j^STD which combines all boat-gear categories:

(7)

The standardized CPUE’s by boat-gear category are obtained by dividing each catch cell C_i,j by the corresponding standardized effort STD E_i,j^STD obtained from (6):

(8)

Lastly the combined standardized catch-per-unit-effort effort in a period j is calculated. Here the combined monthly catch of all boat gear categories is divided by the combined monthly standardized effort obtained from (7).

Catch-Per-Unit-Effort: j=1…12n. (9)

At this stage tasks (a) and (b) that was set-up at the beginning of this section have been achieved.

Consistency issues and need for normalization

Two points arise now regarding:

1. consistency of standardized data and

2. their numerical treatment across different periods.

It is evident that the standardization factors formulated by the presented approach depend directly on the selection of a hypothetical boat-gear category to be used as standard. According to Robson [2] the role of such a standard can also be played by any of the actual boat gear categories, which would result in a different but equally valid set of standardization factors [13,14]. Given that in studying the fluctuation and trend of standardized variables users require consistent sets of data, it becomes apparent that the standardization effort and CPUE values so far obtained need additional treatment in order to become independent of the initial selection of a boat-gear as standard. One way of achieving this is to adopt the normalization approach that was used by the ICES North Sea Round Fish Working Group (1980). The approach consists of (i) calculating the arithmetic mean of a standardized variable across periods and, (ii) substituting each standardized value by its proportion to the mean. In such a manner the resulting normalized values are dimensionless and share a similar value scale [15,16].

It remains to be seen if such normalized values are independent of the choice of a boat-gear category as standard. This is rather easy to prove without performing tedious computations. Suffice to notice that all expressions involving standardized effort contain two factors: one which is the quotient and another that is independent of and depends only on the original data. Consider for instance expression (7) which computes the combined standardized effort for a given period j. By recalling that each and that , this expression can also be written as:

(10)

When the combined monthly standardized effort is summed across periods, its arithmetic mean will also contain . During normalization each standardized effort from (10) will be divided by the arithmetic mean thus canceling out and making the obtained normalized effort independent of the initial choice of a boat-gear category as standard.

Working in a similar manner with the standardized CPUE’s we find that their sums and arithmetic means contain an expression of and other expressions that are independent of it. During the normalization process the expressions of cancel out thus proving that the normalized CPUE’s are independent of the initial choice of a boat-gear category as standard.

Numerical example

Table 2 shows the results of the standardization approach suggested by this study after it has applied to the NFIS catch/effort data of Table 1.

Here the standardization involves m=4 boat-gear categories and 12 catch/effort monthly estimates resulting a total of 48 CPUE’s. It is recalled that during the standardization phase the notation of these CPUE’s will temporarily be changed to P since they will be viewed as representing daily catches. Accordingly their units will be in Kg.

Calculation of standardization factors

First the sum of all 48 daily yields (12 yields for each of the 4 boatgear categories) is calculated.

The corresponding arithmetic mean in (2) will be equal to 6,366/48 = 132.6 Kg.

Next step is the calculation of average daily yields for each boatgear category using expressions (3) and (4).

Launches with traps: P₁ =3,038.5 Kg and

Launches with kingfish net: P₂ = 2,107.5 Kg and

Launches with misc. gear: P₃ = 383.8 Kg and

Speedboats with misc. gear: P₄ = 836.2 Kg and

Calculation of standardization factors (STD) makes use of expression (5). Each of the above averages is divided by calculated earlier:

STD factor for launches with traps = 253.2/132.6 = 1.909.

STD factor for launches with kingfish net= 175.6/132.6 = 1.324.

STD factor for launches with misc. gear = 32.0/132.6 = 0.241.

STD factor for speedboats with misc. gear = 69.7/132.6 = 0.525.

These results are shown in the first block of Table 2. To be noted that once these factors have been calculated they apply to all 12 monthly columns of 2014.

Calculation of standardized effort:

The second block of Table 2 illustrates standardized effort for each of the four boat-gear categories. All standardized effort figures by boatgear category are resulting from the application of expression (6) to all effort cells in Table 1. For instance in January 2014 the actual effort of launches with traps is 3,168 boat-gear days. The standardization factor for this boat-gear category is 1.909. By multiplying the 3,168 actual boat-gear days by this factor we obtain a standardized effort of 6,048 boat-gear days (first cell of the second block in Table 2).

To be noted that since all standardized effort values are addable it is now possible to combine them vertically across boat-gear categories and then horizontally across months, thus obtaining a total effort figure for 2014 equal to 122,733 standardized boat-gear days.

Next line shows combined standardized effort in normalized form. The arithmetic mean of the 12 effort figures is 10,228 boat-gear days. The normalized value of the first entry is 9,934/10,228 = 0.971.The rest of the normalized effort values are calculated likewise.

Calculation of standardized CPUE’s:

The third block of Table 2 illustrates standardized CPUE’s for each of the four boat-gear categories. All figures are resulting from the application of expression (8) to each CPUE cell in Table 1. For instance in January 2014 the standardized CPUE for launches with traps will be 697,000 Kg of catch (first cell in Table 1) divided by the corresponding standardized effort of 6,048 boat-gear days, which gives 115.2 Kg/boat gear day.

A combined standardized CPUE is also computed using expression (9). Here the total catch for January 2014 is 1,281,000 Kg and the combined standardized effort is 9,934 boat-gear days, thus resulting a combined standardized CPUE of 128.9 Kg/boat-gear day.

Next line shows combined standardized CPUE in normalized form. The arithmetic mean of the 12 combined CPUE’s figures is 131.7. The normalized value of the first entry is 128.9/131.7=0.979. The rest of the normalized effort values are calculated likewise.

To be noted that the notation for catch-per-unit-effort has returned back to CPUE since this variable is again calculated as a weighted average of catch divided by effort.

Figure 3 illustrates a plot of the normalized effort and CPUE contained in Table 2.

Results

Application of effort standardization to the fishery of Spangled emperor (Lethrinus nebulosus)

As already mentioned in Introduction Spangled emperor (Lethrinus nebulosus) was the top species in 2014 with landings representing 16.2% of the total.

This species is targeted by launches with traps and speedboats (tarads). Catches by the other two boat-gear categories are negligible and regarded as accidental (Figure 2). Consequently effort standardization focuses on the above two boat-gear categories. Launches with traps are the predominant boat-gear accounting for 76% of the species catches in 2013 and 71% in 2014.

Figure 2: Narrow-barred Spanish mackerel (Scomberomorus commerson).

Table 3 illustrates catch/effort data for 2013 and 2014. Since the effort exerted by the two boat-gear categories is not compatible no combined data are shown for effort and CPUE’s in the last two columns.

Table 3: Catch/effort data for Sh'ari Lethrinus nebulosus (Spangled emperor) (2013 – 2014). Accuracy of estimates: 90.6%.

Table 4 shows the results of the standardization process, including normalized values for effort and CPUE.

Table 4: Standardized effort and CPUE for Sh'ari Lethrinus nebulosus (Spangled emperor) (2013-2014).

Figure 4 illustrates monthly plots of normalized effort and CPUE. There is a slight (but visible) rising trend for fishing effort and a declining one for the CPUE.

Figure 3: Plot of normalized effort and CPUE based on the 2014 NFIS catch/effort data.

Figure 4: Monthly plots of normalized effort and CPUE for Sh'ari Ù?رعش Lethrinus nebulosus (Spangled emperor) (2013-2014). There is a slight (but visible) rising trend for fishing effort and a declining one for the CPUE.

Application of effort standardization to the fishery of Narrow-barred Spanish mackerel (Scomberomorus commerson)

This important species (second in the 2014 ranked landings and representing 10.5% of the total) is targeted by launches with kingfish net and speedboats (tarads). Catches by launches with miscellaneous gear are negligible and are not included in the case study. Launches with kingfish net are by far the predominant boat-gear accounting for 90% of the species catches in 2013 and 95% in 2014.

Table 5 illustrates catch/effort data for 2013 and 2014. Since the effort exerted by the two boat-gear categories is not compatible no combined data are shown for effort and CPUE’s in the last two columns.

Table 5: Catch/effort data for Narrow-barred Spanish mackerel (Scomberomorus commerson) (2013-2014). Accuracy of estimates: 88.3%.

Table 6 shows the results of the standardization process, including normalized values for effort and CPUE.

Table 6: Standardized effort and CPUE for Narrow-barred Spanish mackerel (Scomberomorus commerson) (2013-2014).

Figure 5 illustrates monthly plots of normalized effort and CPUE. There is a slight (but visible) declining trend for both fishing effort and the CPUE.

Figure 5: Monthly plots of normalized effort and CPUE for Narrow-barred Spanish mackerel (Scomberomorus commerson) (2013-2014). There is a slight (but visible) declining trend for both fishing effort and the CPUE

Discussion

Comparison to other methods

As mentioned in the Introduction the National Fisheries Information System (NFIS) has recently adopted the presented approach that combines elements of the normalized relative effort (used by the North Sea Round Fish Working Group, ICES, 1980) and the relative fishing power developed by Robson [2]. It was also mentioned that although the existing literature offers a plethora of other more recent and more sophisticated methods it was nevertheless considered preferable to first try out approaches that (a) depend only on catch/effort data from commercial fisheries and, (b) are applicable to situations of limited time coverage.

The Robson basic concept of relative fishing power was adopted in formulating effort standardization factors as shown in expressions (1) - (5). The presented study uses a variation to the Robson concept; instead of arbitrarily selecting an existing CPUE to use as standard it uses for this purpose a mean daily yield of one fishing unit of a hypothetical boat-gear category. This variation does not constitute a real difference since Robson states that in choosing a CPUE standard “any boat-gear is as good as another”. It is the authors’ view, however, that involving all boat-gear categories in the source data makes the selection of the CPUE standard less arbitrary.

On the other hand the fact remains that users should be free to use any standard that would be appropriate or convenient for their work. This means that several standardized datasets, all equally valid but different from each other, might be resulting from the same source data. To overcome this problem the presented method further processes the standardized data with the objective of making them consistent irrespective of the initial choice of a CPUE as standard. It was shown that such an objective can be achieved by means of a normalization process such as the one adopted by the North Sea Round Fish Working Group, ICES (1980).

Lastly the presented method follows the same concept of dynamic standardization shown in both ICES and Robson approaches. Monthly and annual standardization factors (and hence normalized effort and CPUE’s) vary when the source data cover different numbers of years. For instance, launches with kingfish net have a standardization factor of 1.840 over the period January 2013-December 2014. This value will be different when the source data will extend to December 2015, December 2016, etc. Such a consideration is essential in order for the standardized variables to be compatible across all periods, a criterion that would not be met if standardization was to apply for each year separately.

Equivalent approaches for the formulation of standardization factors

Expression (5) in Section 2 specifies that the standardization factor for a specific boat-gear category is directly defined as the ratio of its overall CPUE (viewed temporarily as the average daily yield of a single fishing unit over all periods) to the average daily yield of a hypothetical boat-gear. It is recalled that i P is obtained from expression (4) and from expressions (1) and (2).

The chosen approach however does not preclude the adoption of other hypotheses which can produce the same results by means of different interpretations of the CPUE’s. For instance an alternative approach is to formulate standardization factors on the basis of days needed catching the same arbitrary quantity Q. Under such a scheme the days needed for each boat-gear to catch Q will be Q/ Next a hypothetical boat-gear category with catch-per-unit effort equal to is considered. Here the number of days needed to catch Q is equal to . Since the number of days needed is in reverse proportion to the relative importance of a boat-gear (i.e., higher performance implies fewer days to catch a given quantity Q) we divide the second ratio by the first, thus obtaining the same standardization factor .

The problem of data gaps

Maunder [1] has stressed the importance of paying due attention to situations in which there are data gaps in the datasets. The remedies are not always simple and in some cases they become quite elaborate.

It is the authors’ view that the problem of data gaps does not affect the presented method since the standardization factors are calculated on the basis of cumulative daily yields covering the entire reference period. It was shown that the standardization process applies to a matrix of source data (as shown in Tables 1,3 and 5) in which data cells may as well contain zeroes (for instance the speedboats in February and August 2013 and in February 2014). In mathematical terms the only condition for a boat-gear category to participate in the process is to have at least one non-zero entry in the matrix. In practice, however, boat-gear categories showing small and scattered quantities of accidental catch are not included in the process as was for instance the case or launches with miscellaneous gear catching kingfish.

Another point worth addressing is the reliability of catch/effort estimates that constitute the data source for the standardization process.

In Qatar the NFIS catch/effort are collected in conformance to strict norms concerning sample size and frequency of sampling. Raw data go through a gauntlet of various quality checks before they are processed and the resulting estimates are subject to quality checks relating to accuracy. The aim of such rigorous monitoring is to achieve a compound accuracy of catch/effort estimates that stays above 90%; this has been consistently achieved from 2014 onwards. In Qatar use is made of the “pessimistic” accuracy concept in which the resulting accuracy stays above a pre-set lower limit [17-19]. It is also a composite index incorporating a spatial accuracy (a function of sample size) and a temporal accuracy (that depends on sampling frequency). In addition to the above two relative indices of accuracy the Sampling Uniformity Index (SUI) monitors the uniformity of samples over the sampling days and it penalizes the temporal accuracy in cases of uneven concentrations of samples favouring certain sampling days.

References

1. Maunder MN, Langley AD (2004) Integrating the standardization of catch-per-unit-of-effort into stock assessment models: testing a population dynamics model and using multiple data types. Fish Res 70: 385-391.
2. Robson DS (1966) Estimation of the relative fishing power of individual ships. ICNAF Res Bull 3: 5-14.
3. Gavaris S (2011) Use of a multiplicative model to estimate catch rate and effort from commercial data. Can J Fish Aquat Sci 3: 2272-2275.
4. Ortiz de Zárate V, Ortiz de Urbina JM (2008) Standardized fishing effort of the Spanish baitboat fleet targeting Albacore (Thunnusalalunga) in the Northeast Atlantic, 1981-2005. Collect Vol Sci Pap ICCAT 62: 816-823.
5. Anticamaraa JA, Watson R, Gelchu A, et al. (2011) Global fishing effort (1950–2010): Trends, gaps, and implications. Fisheries Research 107: 131-136.
6. Campbell RA (2004) CPUE standardization and the construction of indices of stock abundance in a spatially varying fishery using general linear models. Fish Res 70: 209-227.
7. Francis RICC (1999) The impact of correlations in standardized CPUE indices. NZ Fisheries Association Research Document No. 99/42.
8. Gulland JA (1956) On the fishing effort in English demersal fisheries. Fishery Investigations Series II. Ministry Agriculture, Fisheries and Food, London 22: 45.
9. Hansen KE, Ferro RST, Hopper A, et al. (1998) In: Barthel KG et al. (eds) Investigation of the relative fishing effort exerted by towed demersal gears on North Sea consumption species. Proceedings of the third conference of European marine science and technology 5: 148-149.
10. Harley SJ, Myers RA, Dunn A (2001) Is catch-per-unit-effort proportional to abundance? Can J Fish AquatSci 58: 1760-1772.
11. Hinton MG, Maunder MN (2003) Methods for standardizing CPUE and how to select among them. ICCAT Document No. SCRS/2003/034.
12. Maunder MN, Starr PJ (2003) Fitting fisheries models to standardized CPUE abundance indices. Fish Res 63: 43-50.
13. Pearse PH(1980) Regulation of fishing effort. FAO Fisheries Technical Paper No. 197 (FIPL/T197). Food and Agriculture Organization of the United Nations, Rome 82.
14. Porter JM, Oritz M, Paul SD (2003) Updated standardized CPUE indices for Canadian bluefin tuna fisheries based on commercial catch rates. ICCAT Col VolSci Pap 55: 1005-1018.
15. Quinn TJ, Hoag SH, Southward GM (1982) Comparison of two methods of combining catch-per-unit-effort data from geographic regions. Can J Fish AquatSci 39: 837-846.
16. Tidd AN (2013) Effective fishing effort indicators and their application to spatial management of mixed demersal fisheries. Fisheries Management and Ecology.
17. Stamatopoulos C (1999) Observations on the geometrical properties of accuracy growth in sampling with finite populations. FAO Fisheries Technical Paper No. 388.
18. Stamatopoulos C (2002) Sample-based fishery surveys. A technical handbook. FAO Fisheries Technical Paper No. 425.
19. Stamatopoulos C (2003) Safety in sampling. FAO Fisheries Technical Paper No. 454.