ICES PGRFS REPORT 2011 Reports/Expert Group...Denmark Telephone (+45) 33 38 67 00 Telefax (+45) 33 93 42 15 [email protected] Recommended format for purposes of citation: ICES. 2011. Report

ICES PGRFS REPORT 2011 ICES ADVISORY COMMITTEE

ICES CM 2011/ACOM:23

Report of the Planning Group on Recreational Fisheries Surveys (PGRFS)

2-6 May 2011

Esporles, Spain

International Council for the Exploration of the Sea Conseil International pour l’Exploration de la Mer

H. C. Andersens Boulevard 44–46 DK-1553 Copenhagen V Denmark Telephone (+45) 33 38 67 00 Telefax (+45) 33 93 42 15 www.ices.dk [email protected]

Recommended format for purposes of citation:

ICES. 2011. Report of the Planning Group on Recreational Fisheries Surveys (PGRFS) , 2-6 May 2011, Esporles, Spain. ICES CM 2011/ACOM:23. 111 pp.

For permission to reproduce material from this publication, please apply to the Gen-eral Secretary.

The document is a report of an Expert Group under the auspices of the International Council for the Exploration of the Sea and does not necessarily represent the views of the Council.

© 2011 International Council for the Exploration of the Sea

ICES PGRFS REPORT 2011 | i

Contents

Executive Summary ............................................................................................................... 1

1 Opening of the Meeting ............................................................................................... 3

2 Introduction .................................................................................................................... 4

2.1 European context and target species .................................................................. 4

2.2 Terms of reference ................................................................................................ 5

3 Compilation of subgroup reflections ......................................................................... 7

The participants divided into three subgroups to identify issues that we would need to address in designing and conducting surveys of recreational fisheries that would produce reliable estimates of total catch for important fish stocks. ........................................................................... 7

3.1 Sub group 1: Regional issues ............................................................................... 7

3.2 Sub group 2: Common methodology issues ..................................................... 7 3.2.1 Participation rate ...................................................................................... 7 3.2.2 Data on catches ......................................................................................... 8

3.3 Sub group 3: Formulation of recommendations ............................................. 11

3.4 Term of references for the next meeting .......................................................... 12

4 International experiences in applying recreational fisheries sampling schemes .......................................................................................................................... 13

4.1 Denmark .............................................................................................................. 13 4.1.1 Introduction ............................................................................................ 13 4.1.2 Methods ................................................................................................... 16 4.1.3 Results ..................................................................................................... 20 4.1.4 Discussion ............................................................................................... 27 4.1.5 Sources of error ...................................................................................... 29 4.1.6 Fishing without license ......................................................................... 30 4.1.7 Conclusion .............................................................................................. 30 4.1.8 References ............................................................................................... 31

4.2 England ................................................................................................................ 39 4.2.1 New results in 2011 based on methods already described in

the 2010 report. ....................................................................................... 39 4.2.2 New methods tested in 2011. ............................................................... 39

4.3 Finland ................................................................................................................. 42 4.3.1 New results in 2011 based on methods already described in

the 2010 report. ....................................................................................... 42 4.3.2 New methods tested in 2011. ............................................................... 42

4.4 France ................................................................................................................... 43 4.4.1 Possible sampling frames ..................................................................... 43

4.5 Germany .............................................................................................................. 51 4.5.1 General overview of national recreational fisheries .......................... 51

ii | ICES PGRFS REPORT 2011

4.5.2 Detailed description of national recreational fisheries ..................... 51 4.5.3 Sampling frames .................................................................................... 52 4.5.4 Survey methods ..................................................................................... 52 4.5.5 Results ..................................................................................................... 52

4.6 Latvia .................................................................................................................... 63 4.7 Netherlands ......................................................................................................... 65

4.7.1 General overview of national recreational fisheries .......................... 65 4.7.2 Recreational fisheries survey design ................................................... 66 4.7.3 Preliminary results Recreational Fisheries Survey ............................ 67 4.7.4 Recall bias in historical catch estimates .............................................. 72 4.7.5 Economic value Recreational Fisheries ............................................... 72 4.7.6 Dutch recreational fishermen abroad .................................................. 73 4.7.7 Recommendations for the Netherlands .............................................. 74

4.8 Norway ................................................................................................................ 76

4.9 Poland................................................................................................................... 76

4.10 Spain – Basque country (Atlantic area) ............................................................ 76 4.11 Spain – Balearics Islands (NW Mediterranean) .............................................. 76

4.11.1 New results in 2011 based on methods already described in the 2010 report ........................................................................................ 76

4.11.2 New methods tested in 2011 ................................................................ 83

4.12 Sweden ................................................................................................................. 87 4.12.1 Possible sampling frames ..................................................................... 87 4.12.2 Available statistics ................................................................................. 89

4.13 U.S.A 91 4.13.1 Updates on the U.S. Marine Recreational Information

Program ................................................................................................... 91 4.13.2 Matching Estimation Methods with Sampling Designs in

Access Point Surveys ............................................................................. 91 4.13.3 New Access Point Survey Sampling Design ...................................... 96 4.13.4 New Dual-Frame Sampling Designs for Surveys of Fishing

Effort ........................................................................................................ 98 4.13.5 Other Ongoing MRIP Studies ............................................................ 100

Annex 1: List of participants............................................................................................. 103

ICES PGRFS REPORT 2011 1

Executive Summary

PGRFS became WGRFS (Working Group on Recreational Fisheries Surveys). WGRFS discusses, evaluates and develops common methods for recreational fisheries sur-veys. EU Members States (MS) are required to establish such programmes for several key species (cod, European seabass, eels, salmon and bluefin tuna according to ICES areas) in order to meet the requirements of the EU Data Collection Framework (EC Regulations 199/2008 and EC Decision 2008/949/EC).

The first part of the meeting was devoted to country presentations. Each country presented its own context, methods, results and improvements from the last year. The following countries carried out a presentation: Denmark, Norway, Poland, France, Finland, Sweden, Germany, Spain – Basque Country, Spain – Balearic Islands, UK and Latvia.

The IMEDEA Project CONFLICT was presented focusing on the development of a recreational fisheries assessment program in the Balearic islands, from an ecological, economic and social point of view.

Next, Dave Van Voorhees presented the latest results of ongoing projects focused on improving recreational fishing surveys in the U.S.

The following topics were discussed during the meeting:

• For countries that have applied a recall-based interview approach, the importance of accounting for potential non-response bias was discussed and compared between MS.

• The potential of using recreational fisheries surveys to collect data for de-signing descriptors to support the Marine Strategy Framework Directive, aiming at achieving or maintaining a good environmental status by 2020 at the latest, was discussed.

• Several presenters highlighted the need for registries of fishing partici-pants in European countries in order to improve the efficiency of recrea-tional fishing surveys (more cost-effective monitoring). Indeed, without such registries the current monitoring is very inefficient and it can be very expensive to obtain the sample sizes required for estimating fishing effort and/or catch at desired levels of statistical precision. These regis-tries should include contact information (at least a phone number and mailing address). Some registries already exist for Poland, Germany, Finland, Spain and Denmark (mainly based on fishing permit system). But they are most of the time unusable due to lack of important or inade-quate information.

• Depending of the use of data many recreational catch estimates will have to undergo a transformation from numbers to weights and vice versa. There are two reasons for this 1) in commercial fishery the catches are normally given in weight whereas many inland and recreational fisheries are normally given in numbers and 2) some recreational fishers using commercial fish gear tend to recall weight better than abundances, de-pending on the fishery and target species. This transformation requires an average weight estimate, which in many countries was a significant source of error. Obtaining average weight for different species, seasons and waters should be coordinated, to avoid regional data overlap.

2 ICES PGRFS REPORT 2011

• Some comparisons have been carried out regarding the rate of response coming from different surveys. The following surveys were compared to each other:

o Internet and telephone surveys

o telephone and diary surveys

o telephone and mail surveys

• The efficiency of these methods greatly depends on the national context and on the availability of national registries.

The second part of the workshop was dedicated to breakout group discussions about: regional issues, common methodology issues and the formulation of recommenda-tions.

The subgroup working on regional dimensions, suggested the development of a common regional plan and regional reporting, to share and discuss data and estima-tions between countries of the same region before providing them to the EU, with the aim to integrate regional issues regarding recreational fishing in addition to DCF requirements.

The subgroup working on common methodology, suggested the following approach regarding the following two issues: (1) improving participation rate and (2) improv-ing catch data quality. For improving participation, it was suggested to follow a list of specific steps: screening survey to have a first map of recreational fishers with very few data, detailed surveys allowing the collection of more specific data through dif-ferent ways. Catch data can be improved with panel logbook/diaries and with access point surveys, requiring new methods (video cameras, google maps etc.).

The subgroup working on recommendations, prepared a list of eight core issues con-cerning: Registry, International Coordination, Confidentiality, Data Quality, Mortali-ty, Stock Assessment, Species, and Report Format. These recommendations have finally been used as the general recommendations of the working group and are de-scribed in the recommendation section.


1 Opening of the Meeting

The PGRFS meeting took place from 2-6 May 2011, at the IMEDEA laboratory in Esporles (Palma de Mallorca, Spain). The participants at the meeting are listed in Annex 1.


2 Introduction

2.1 European context and target species

The EU Data Collection Framework (EC 199/2008) defines recreational fisheries as “non-commercial fishing activities exploiting living aquatic resources for recreation or sport.” A range of other definitions of recreational fishing are given in Pawson et al. (2008). The scientific assessments of European marine fish stocks continue to focus on quantifying the mortality associated with commercial fishery removals, and have ignored the impacts of recreational fishery catches. However there are species such as European sea bass and cod which are widely targeted by recreational fishermen and where data from recreational fisheries could potentially improve the assessments. Stocks of cod, bluefin tuna and eels (Anguilla anguilla) in European waters are se-verely depleted, and for such stocks, it is clearly important to be able to quantify all sources of fishery removals that could affect recovery. Such factors are presumably the primary reason for the EU Data Collection Framework requirement to quantify recreational fishery catches of these species.

Prior to the requirements of the EU Data Collection Framework and the preceding Data Collection Regulation, studies of recreational fishing in Europe often focused more on descriptions and socio-economic aspects rather than estimating catch quanti-ties using the types of survey approaches used in the United States and elsewhere (see references in Pawson et al. 2008). The EU DCR/DCF requirement for pilot studies to collect the information necessary to establish recreational fishery surveys has re-sulted in a greater focus on the appropriate methodology for estimating catches, and there have also been surveys of freshwater recreational fisheries in Germany using methods such as telephone-diaries (see Section 12). The ICES Planning Group on Commercial Catches, Discards and Biological Sampling (ICES, 2008) considered that a forum was needed to consider appropriate methodology for European fisheries and to promote harmonisation of approaches between countries as far as possible. WKSMRF was therefore established to address the following Terms of Reference in relation to European recreational fisheries:

a) Provide a comprehensive description of the marine recreational fisher-ies in each ICES country including the species/stocks targeted, the po-tential or known magnitude of recreational catches and effort by geographic area, time period and fishing method, and the definition of appropriate reference populations of recreational fishermen for sam-pling;

b) Review the findings of existing studies on recreational fisheries includ-ing DCR Pilot Studies and their relevance for sampling schemes in other areas;

c) Recommend appropriate statistical sampling schemes, protocols, and associated data analysis for estimating recreational fishery removals and length/age compositions, taking account of international experi-ence and recent methodological developments. Review potential for conducting parallel studies to establish comparability of results for dif-ferent sampling schemes. The legal framework for collection of recrea-tional fisheries data by EU Member States is given by the EU Data Collection Framework (Council Regulation (EC) No 199/2008 and Council Decision 2008/949/EC). The Council Decision specifies that:


• For the recreational fisheries targeting the species listed in Appendix IV (1 to 5), Member States shall evaluate the quarterly weight of the catches.

• Where relevant, pilot surveys as referred to in Chapter II B (1) shall be carried out to estimate the importance of the recreational fisheries mentioned in point 3(3)(a).

• Data related to annual estimates of the catches in volumes must lead to a precision of level 1 (level making it possible to estimate a parameter either with a precision of plus or minus 40 % for a 95 % confidence level or a coefficient of variation (CV) of 20 % used as an approximation).

Appendix IV of Council Decision 2008/949/EC specifies fleet metiers covered by the DCF, and includes recreational fisheries specified to Level 5 in the matrix (target spe-cies assemblage). The species for which recreational fishery data are to be collected in each area are:

• Baltic (ICES Sub Divisions 22-32): Salmon, cod and eels • North Sea (ICES Div. IV & VIId) and Eastern Arctic (ICES Div. I & II): cod

and eels • North Atlantic (ICES Div. V-XIV): Salmon, seabass and eels • Mediterranean and Black Sea: bluefin tuna and eels

The recreational fishery data do not have to be collected according to mesh size bands of nets (metier Level 6), but the DCF specifies that data should be collected for “all vessel classes (if any) combined”. The DCF does not specifically mention shore-based (i.e. non-vessel) recreational fishing.

2.2 Terms of reference

The terms of reference for the WGRFS 2011 workshop are the following:

a) Define common definitions and terminology and create a glossary about surveys of recreational fishing.

b) Identify methods to evaluate and compare potential for recall, nonresponse and undercoverage biases in alternative survey designs.

- Develop guidelines for survey design and analysis methods for the estima-tion of recreational fishing effort and catch totals, including the estimation of totals for subpopulations defined by geography (country of fishing ac-cess or water body of fishing), residency status (country resident or tourist), or mode of fishing (shore, private boat, or for-hire boat). Guidelines for minimizing possible biases

- Guidelines for maximizing precision

- Guidelines for analyzing cost-effectiveness and for optimizing survey de-signs under limited budgets

c) Consider alternative designs for surveys of fishing participation, effort, and catch, and assess their utility relative to their potential for minimizing bias, maximizing precision, and controlling costs.

d) Examine how best to utilize alternative survey designs to provide needed segmen-tation of estimated population totals into different metiers through either sample stratification or domain estimation.

e) Develop a communications and education strategy that will promote a greater stakeholder involvement, participation, and support.


f) Define the needs for expert statistical consultant support and develop a plan for acquiring and directing that support to facilitate collaborative development of op-timal survey designs.

g) Motivate regional cooperation (stock estimates)

h) Plan future collaborations among European countries in the development of com-patible survey designs that will provide unbiased and reliable recreational fishery catch statistics for shared fish stocks (milestone for 2013)


3 Compilation of subgroup reflections

The participants divided into three subgroups to identify issues that we would need to address in designing and conducting surveys of recreational fisheries that would produce reliable estimates of total catch for important fish stocks.

3.1 Sub group 1: Regional issues

Take into account other species for having a good description of recreational fisheries and to designate which are more relevant for regional assessment. Monitored species should be those of importance for each country (example of cod in France, seabass in England).

The choice of the targeted species should also be taken into account from different arguments: if stock assessments or management plans are implemented or if this is an endangered species

We consider two scales of study: the national scale, and the regional scale, which could include various countries of a same area or sharing common cases of study.

Having a regional approach is valuable for the description of the biological parame-ters (evaluated at sub-population scale) but is not for the socio-economic, which could be more national based. Thus, we would propose separate surveys on i) socio-economic aspect and ii) biological ones.

We think that it is important to have a common investigation for species for which it is possible to get DCF funding.

It could be interesting to develop a regional common plan and then regional report-ing for different countries

It should be also necessary to share and discuss data and the estimations between countries of a same region before providing them to EU.

The priority and standards for information to collect should be taken into account, but the DCF mentioned only catches quarterly. It may be possible to take into account a limited list of recommendation for each of the member countries

3.2 Sub group 2: Common methodology issues

3.2.1 Participation rate

Screener Survey (mail or phone) -

The first contact could be limited to determining if any household residents partici-pate in recreational fishing, collecting their contact information, and recruiting them for participation in a more detailed follow-up survey. Sweden has used this approach and have conducted a screener survey every second year now (via mail). It is planned to have a screener every fifth year instead and a more detailed survey annually. In the first contact, the main goal is to identify voluntary anglers that would agree to be contacted again for the more detailed survey.

Detailed Survey –

If possible, lists of vessels, anglers, and businesses have to be used. A dual-frame sampling approach like the one being tested in the US could be used, where both the


mailing address list and the license list are used. Comparison of the response rates for mail and telephone contacts could be evaluated by drawing two independent sam-ples and using the phone numbers of the listed persons selected in the second sam-ple.

Questionnaire:

As general recommendations, we propose to ask as few questions as possible and to leave out the questions that may be provocative (e.g. income), thus including only the relevant questions (which could vary depending on the context: e.g. it is important to ask about fishing from the second-home in Scandinavia). The necessary questions identified there to allow to segment the angler population could be: information on the urban-rural area, age, region (coast vs. inland), gear, fishing environment, type of boat if used, targeted species, fishing period, avidity (nb. of fishing trips/days/hours), years of fishing experience etc…

Contact period:

Contact periods and recall periods will have to be adjusted to the structure of the fishery in any country (e.g. contact them in the fall if the summer is the main fishing period).

Independent vs. longitudinal sampling:

It is possible to use a series of independent surveys, a panel survey approach that re-contacts the same individuals over a period of time (contacting the participants on a certain time interval), or some combination of both (e.g., in a rotating panel design) for a detailed survey.

A panel survey could be conducted with or without a diary, or logbook, but the use of such tools as “memory joggers” should improve the recall accuracy of respon-dents.

An internet approach could also be used, where panel members are asked to report about their fishing trips on a website form. However, the possibilities of setting up a web survey and applying email for contact varies between countries and there are risks for bias (e.g. elderly people cannot be approached through Internet).

3.2.2 Data on catches

3.2.2.1 Access points:

Visual surveys can be used through video cameras in boats, harbours or planes. For example, Google Maps have been used in Spain to identify the number of angling boats based on visible fishing gears.

In Swedish harbours, the scientists are using maps and have anglers plot into the map the number of fishing trips in each location. This information is then used in combination with the mailing address data for a given respondent to evaluate the distance between the residence and the fishing area (app. 30 km). Sweden also asks the anglers if they go abroad to fish.

3.2.2.2 Panels – logbooks/diaries:

It could be difficult to retain anglers in a panel for an extended period of time. We suggest rotating the participants in panels. (logbooks can be on number of fishing


trips only – that may be enough if you get the catch rate from on-site survey). Austra-lia has a four year panel period. This retention period may prove to be too long in other countries, and many panel members may drop out before completing the planned study. Therefore, it may be better to rotate overlapping panels. For exam-ple, a new panel could be recruited every 2 months and asked to report over a 6 month period. In this case, successive panels would be partly overlapping such that three independent panels would simultaneously be reporting data for any given two-month period. Such a design could potentially be more expensive to implement, but it could also provide a larger effective sample size with less risk of possible biases caused by panel members dropping out.

Other solutions for increasing the response rate could be proposed: gifts, lottery, giv-ing back information (on forums, mail results back, have stakeholders doing the sur-vey’s promotion etc.).

Lists for on-site and access point survey sample frames:

Tourism businesses

Harbours

Airport

Ferry terminals

At boarder

Ramp sites

Slip ways

On-site sampling surveys are important as a means of validating the data acquired through off-site surveys. It is important to include all specific locations with fishing participants in the sample frame. A probability-based sampling approach can be used that would give the more frequently visited locations a greater chance of being selected, but the frame should not be restricted to those high activity sites.

Lists for off-site sample frames:

Boat registry

Vessel license registry (charter, head or for hire-boats)

Licence registry

Tourism businesses

Rental homes with a boat (Norway)

It is important to make sure that the surveys are based on a set of lists that are com-prehensive, as well as representative (e.g. license holders may be the more avid fish-ers, but there may be less avid fishers who fish without a license).

Precision: In general, the more people you contact the better the precision you will get on your statistical estimates of fishing effort and/or catch.

Bias: In general, the risk of possible biases will be greatly reduced by providing more complete coverage of fishing participants (less chance of errors due to differences between on-frame and off-frame participants), by obtaining higher response rates (less chance of errors due to differences between respondents and non-respondents),


and by obtaining more accurate responses (less chance of recall or measurement er-rors).

STAKEHOLDER INVOLVEMENT – It is very important to involve stakeholders in the planning of surveys and inform them about how the surveys are designed to pro-duce reliable results.

Why?:

1. To provide legitimacy for the studies

2. To assist in survey design/redesign and design questionnaires,

3. Assist in data collection (motivate anglers to assist in data collection).

How?:

1. Panels for data collection

2. Reference groups/committees,

3. Provide information (in the committee groups, to the panels, at confer-ences/meetings/fairs, through distributing fact sheets/leaflets (via mail, web-sites, meetings), websites, journals/newspapers and other media.


3.3 Sub group 3: Formulation of recommendations

1. Introducing mandatory fishing license

1. Most MS have no mandatory fishing licensing system in place. A mandatory fishing licensing system should include identification of license holders. When appropriate this should be the postal mailing address, mobile telephone number and/or social security number. Possibly fishing licenses could be separated into modalities, e.g. angling, passive gear, spear fishing. MS to provide concrete information on the existing national licensing system prior to next years meeting.

2. Setting up national registry of leisure vessels

2. Knowledge on the number of leisure boats is lacking in most MS.

International coordination

1. Tourist fishermen 1. Organize a workshop on tourist fishermen activities.

General common methods

2. Provide annual estimates

a. Evaluate the potential of MS sampling frames

3. Panel surveys (Method Group)

b. Develop common sampling frames

c. Define tourist fishing variables/modalities

2. Recreational fishing estimates should be provided annually. This is in conflict with the current DCF that requires quarterly estimates. Finer temporal stratification adapted to MS circumstances (e.g. quarterly or by season) can improve precision of annual estimates when appropriate.

1. Provide legal expertise to solve national and regional confidentiality issues.Action: Invite expert to next PGRFS meeting.

1. Quantifying uncertainty (bias and precision)

1. Develop score card with bias indicator for recreational catch estimates. Reference: WKACCU (Workshop on Methods to Evaluate and Estimate the Accuracy and Bias ).

2. Catch versus Landings

Action: Organize break out group at next year’s meeting to develop score card (2 days).2. Agree on a common definition of terminology in recreational fishing.Agreed Definition: Total catch in numbers = numbers of fish kept (harvest) + number of fish released (dead or alive).

1. Emphasis on quantification of release and release mortality (survival rate).Action: Pilot study on release mortality

Licensing / Registry

Confidentiality / Privacy

1. Granting access to individual data

ICES PGRFS 2011 Recommendations (1/2)

Uncertainty / Data quality

Mortality 1. Catch & Release, Mortality


1. Include recreational data in fish stock assessments, management plan evaluations, LTMP or equivalent.

Action: Harry join next year’s Baltic co stock assessment.

Martin participate at WGEEL; MS contact national correspondents.1. Some species (not covered by the DCF) have considerable recreational fisheries catches (or red list species) and should be sampled in the future.Action: Harry contact national correspondent to evaluate communication channels to EU Commission.

1. Develop and agree on common format of national reports according to ICES …. All MS to provide national reports in common format prior to next year’s meeting.

Action: Martin to formulate concrete guidelines for national report structure.

Stock assessment

1. Taking recreational fisheries into account for stock management

ICES PGRFS 2011 Recommendations (2/2)

Species 1. New species: Sea trout, European lobster,

Report format 1. Common format of National Reports

3.4 Term of references for the next meeting

The terms of reference for the next Working Group on Recreational Fisheries Surveys are the following:

a ) Develop and implement a score card system (see for example: WKACCU - Workshop on Methods to Evaluate and Estimate the Accuracy and Bias) in order to evaluate country survey programs

b ) Provide a mini-review on mortality and catch and release (with a method replicable for different species.

c ) Provide a mini glossary on recreational fishing terminology. d ) Address relationships between scientists and stakeholders (association of

recreational fishers, commercial world, administration in charge of regu-late recreational fishing activities)

e ) Discuss the use of the data provided (invitation of someone of DG MARE for addressing this point.

The next meeting of the PGRFS is planned for Denmark or Balearic Islands 7-11 May 2012.

New elected chairmen are Claus Reedtz Sparrevohn (Denmark) and Harry Vincent Strehlow (Germany)


4 International experiences in applying recreational fisheries sampling schemes

4.1 Denmark

Abstract Marine recreational fishing is a popular outdoor leisure activity but the impact on the targeted stocks is often unknown. In order to estimate cod, eel and sea running trout harvest (fish caught and kept) in the Danish angling and passive gear fishing two interview surveys were conducted in January 2011. Recreational fishing was sepa-rated into anglers (with rod and reel) and passive gear fishing (fyke – and gillnets). In 2010 a total of 152,723 anglers and 33,734 passive gear fishermen had issued the com-pulsory annual license. In total, it was estimated that 116 t (RSE=10%) eel, 1,666 t (RSE=6%) cod and 600 t (RSE=%) seatrout (including freshwater catches) was har-vested in the recreational fishery. Eel are almost exclusively taken in the passive gear fykenet fishery and seatrout was mainly caught by anglers which accounted for 90 % of the total harvest. The estimated cod harvest was also mainly by anglers and at least two areas were identified with a high recreational harvest relative to the total yield (commercial landings plus recreational harvest), i.e. the Sound and in Kattegat. Pre-sent interview survey indicates that approximately 5.7% of the total Danish cod yield and approximately 22 % of the total eel yield are taken in the recreational fishing. In the estimation, harvest taken by fishermen without a legal license was included, which increase the estimated harvest with 17% and 24 %, respectively for passive gear and angling.

4.1.1 Introduction

4.1.1.1 Monitoring of recreational fishing

Within Europe the management of recreational fishing has so far mainly been con-ducted on a national level without including catches in neither stock assessment nor ecosystem based management (Lewin et al.2006; Pawson et al., 2008). However, fish-ing mortality has in some areas been estimated to be comparable to - or even exceed-ing – the mortality caused by the commercial fishery (e.g. Coleman et al., 2004; Morales-Nin et. al., 2005). Therefore there are in many EU member states there is an increasing awareness about the impact caused by this type of fishing (Lewin et al. 2006). As a consequence the EU Council has since 2008, as a part of the Common Fi-sheries Policy, obliged member states to estimate the harvest (those fish caught and retained) taken by recreational fishing (EU Council regulation No. 199/2008). Due to this obligation, Denmark has since 2009 initiated a recall survey to estimate quarterly harvest of cod Gadus morhua, eel Anguilla anguilla and since 2010 sea run brown trout Salmo trutta (seatrout). This report contains the 2010 results of the Danish recreational fishery for these three species.

4.1.1.2 Method approach

In most European member states information on harvest taken in the recreational fishing is gathered using some sort of recall survey (ICES 2010a). A recall survey is a type of off-site survey which relies on collecting information through mail, telephone or internet interviews where respondents are asked to recall e.g. their catches, fishing pattern or number of days fished, within a specific timeframe. A specific problem


with recall surveys is that the longer the timeframe respondents have to recall, the more the results tend to be biased (e.g. Tarrant et al., 1993).

In September 2009, Statistic Denmark and DTU Aqua developed a concept for a com-bined telephone and internet recall survey (See Sparrevohn and Storr-Paulsen 2010, 2011, for further information). This recall survey was further extended in 2010. Ini-tially, one license list interview round was carried out in February 2010 to cover all the catches in 2009. In 2010 this design was improved by conducting two surveys thereby limiting the recall period to a maximum of 6 months. Further, the surveys covering the 2010 catches did also include the harvest of sea trout and the number released.

The interview survey presented in this report, has same structure as in 2010, and is separated into two different phases with their own questionnaires and group of re-spondents: 1) The Omnibus and 2) License holders. The omnibus was only conducted once in 2010 as the results from this interview are not likely to change much. The license list survey was conducted twice covering the two periods from January to June and July to December 2010.

4.1.1.3 Recreational fishing in Denmark

Approximately 5.5 million people reside in Denmark; 2.5 million on the mainland and the rest on islands (source: Statistic Denmark, www.dst.dk). Denmark has a very extensive coastline being 7,013 km long and no citizen lives more than 50 km from the nearest coast (Agerskov and Bisgaard, 2011). Recreational fishing in marine waters is therefore an important national outdoor leisure activity. In 1997, 16.5% of the Danish public considered themselves anglers and 12.5 % claimed to have been fishing within the last year (Bohn and Roth, 1997). Further, it was found that out of those that fished, 25% fished in streams, 30 % in lakes, 27% in put & take ponds, but the majority, 73%, answered marine waters. An economic validation of the recreational fishery under-lines the importance of recreational fishery in Denmark, as it was found that Danish willingness to pay for fishing is among the highest in Nordic countries (Roth et al., 2001; Toivonen et al., 2004).

Recreational fishing in Danish coastal waters differs from what is observed in many other countries, especially outside of Europe, in the sense that two major and very different categories of fishing can be identified. The first one is referred to as passive gear fishing throughout this rapport. Passive gear fishing is carried out using station-ary gear such as gillnets and fykenets. The second category of leisure fishing in salt-water is angling.


Table 1. Number of annual angler- and passive gear licenses issued annually. In 2004 no data was available.

Anglers Passive gear

1999 150526 33575

2000 151529 31709

2001 156769 33715

2002 150925 33888

2003 152534 33516

2005 160942 33430

2006 156474 34277

2007 160664 33787

2008 160186 35221

2009 157939 34565

2010 152723 33734

Anglers - domestic as well as tourists - between 18 and 65 years of age have to pur-chase a license costing 19 € for one year, 13 € for one week and 5 € for one day. All passive gear fishers have to pay a license costing 37 € per year and you are not al-lowed to fish before the age of 12. There are three legal reasons for anglers not to hold a license: 1) persons younger than 18 years or older than 65 years, 2) landowners fish-ing in their private waters and 3) put & take angling.

4.1.1.3.1 Passive gear fishing

Passive gear fishing covers fishing which is carried out using gear such as fykenets and gillnets. For the last 10 years there has on average been 33,700 licenses issued per year (Table 1). The fishery is leisure based and it is illegal to sell the catch. There are restrictions to the effort, as it is only allowed to fish with a maximum of either 3 gill-nets plus 3 fykenets or a total of 6 fykenets. The maximum length of gillnets are 45 m and they are not allowed to be closer than a 100 m from the coastline; a restriction mainly set up to protect seatrout. Further, there are several protected areas such as areas around river mouths, where net are illegal. The gear is typically deployed from a small boat with a limited activity radius, which in practice makes this type of fish-ing more or less stationary.

The main targeted species are eel caught in fykenets and flounder Pleuronectes flesus caught in gillnets (Sparrevohn et al., 2009). It is a traditional fishery that has been practiced for centuries in the coastal areas. Earlier, a recreational fishery using eel-trawl and long-lines was also practiced but eel-trawl is now prohibited and long-line catches are limited. Cod and sea running trout are caught both with gillnets and fykenets in the passive gear recreational fishery, but the catches are believed to be restricted to certain areas (Sparrevohn et al., 2009).

4.1.1.3.2 Angling

Angling in saltwater are carried out in waders along the coastline or from structures such as peers, bridges or with boats as a platform. The majority of anglers (73%) are fishing in marine waters (Bohn and Roth, 1997). During the last 10 years the number of annual licenses issued per year has on average been 155,600 (Table 1). The number


of weekly license issued in 2010 was 17,305 and for daily license a total of 23,716 li-censes were issued. There are no restrictions, e.g. bag-limit, to angling fishery in salt-water besides those that apply to fishing in general, i.e. closed areas, minimum landing size etc. The only exception is that trolling closer that 100 m from the coast-line is prohibited. The main target species in saltwater is seatrout, but garfish Belone belone and cod are also regularly caught as well as salmon and various flatfish spe-cies (Rasmussen and Geertz-Hansen, 2001). Seatrout is besides also caught in fresh-waters.

Baltic salmon is almost exclusively caught angling from medium sized (15-25 ft) boats around the island of Bornholm. Peak periods are during the spring/early summer and October/November. Down rigging is the dominating fishing method.

Platforms used when targeting cod range from beach fishing with rod and reel using casting lures to deepwater jigging onboard chartered boats many miles offshore. There is also a substantial fishing focusing on wrecks. Angling for cod on board pri-vate boats is in addition very popular and could probably accounts for a large frac-tion of the total harvest, at least locally.

4.1.2 Methods

A combined telephone and internet survey based upon two questionnaires, the “Om-nibus” and the “License list”, were developed by Statistic Denmark and DTU Aqua. The interviews were conducted by Statistic Denmark as they hold the expertise in this form of surveys. The questionnaire was prior to the 2009 interview tested upon a subgroup of fishermen, to optimize the process and reformulating questions that potentially could lead to misunderstandings. DTU Aqua was responsible for the fol-lowing data processing.

4.1.2.1 The Omnibus interview

The Omnibus is a monthly interview survey conducted by Statistic Denmark wherein questions are asked on behalf of e.g. companies, newspapers and research institutes. In 2009 three telephone interview rounds were conducted were questions on recrea-tional fishery were included and in 2010 one additional omnibus survey was con-ducted in March. The recreational fishery questions were embedded as a minor part of this interview, hence the nonresponse bias is expected to be insignificant. Respon-dents were selected by telephoning a random number. The interview was conducted with that person within the household who last had birthday. Only citizens between 16 and 74 were included. A total of 958, 957 and 968 were interviewed and answered in 2009 and in March 2010 a total of 985 were interviewed. The main objective was to estimate the population size that fished without a license and their effort.


Figure 1. Area definition used in the interview survey.

Fishermen not holding a license were asked for their reasons. There are several legal exemptions from the compensatory license for angling fishing (see section 1.3). Pas-sive gear fishers do not have any legal excuse for not holding a license when fishing in saltwater.

Furthermore, respondents were asked for information on effort in fishing days to be able to estimate whether people fishing without a license are fishing with same effort as people holding a license. These questions provided the needed information for calculating the fraction of illegal fishermen and the effort they fished with. Respon-dents were also asked about their fishing pattern outside Denmark, such as countries they had visited for fishing.

4.1.2.2 The License interview

This recall survey targeted fishermen with a valid annual license at the time of the interview. The main difference between the 2009 and 2010 “license list” question-naires was 1) that seatrout harvest was included 2) questions about released fish asked and 3) two interview rounds 6 months apart was conducted. In order to esti-mate the 2009 harvest only one interview round was carried out in 2010, which means that in the maximum recall period was 12 months for the 2009 data whereas this was only 6 month in 2010. The data in this report are based upon interview rounds that were conducted in July 2010 and in February 2011. Since two different license lists are available, one for anglers and one for passive gear fishermen, there was conducted two surveys with quite identical questionnaires. Independent of list, the respondent was randomly selected and initially contacted by letter wherein they were encouraged to answer the questions via the internet. If no respond was detected after a period of around two weeks, the respondent was contacted by telephone and - if reached - encouraged to answer via the internet or via telephone. This question-


naire contained detailed questions on species harvested, released and fishing effort within the last 6 months. The respondent was explicitly told to distinguish between the part of the catch kept (i.e. the harvest) and the part discarded/released. To esti-mate harvest by ICES managing areas (Fig. 1) and quarter the respondents were asked to provide the information per area and quarter. In the Danish license system it is also possible to issue a license valid for one day or one week. However, the number issued of these licenses is relatively small compared to the number of annual licenses. Therefore, no separate interview was conducted for these two groups. However they were accounted for in the total harvest estimation.

Purchasing a license for passive gear fishing automatically gives license to angle with rod and reel as well. To include this group in the estimates, all passive gear fishermen were asked whether he/she also angled, a group referred to as “angling with a pas-sive gear license”. An additional interview was therefore conducted on this group in order to estimate their harvest when angling.

4.1.2.2.1 Telephone contra intranet

An analyze was conducted where the total estimated harvest was calculated based upon either 1) those respondents that had responded via the internet or 2) those that was later contacted by telephone and answered there. The motivation for conducting this analyze was: if no differences was found it would not be necessary to contact respondents via telephone in the future, which is the most costly part of the interview round.

4.1.2.3 Analytical methods

Estimating the total harvest or numbers released of cod, seatrout and eel in the Dan-ish recreational fishing was done by estimating the harvest on basis of the reported catches from the license list recall survey. These values were then extrapolated to the entire population of fishermen (all license holders and fishermen without a license) using the effort information collected during the omnibus survey. Different effort levels for those fishing without a license, on a weekly or on a daily license was ac-counted for in the calculation. To compute the total harvest or released numbers Yijof either cod, seatrout or eel per quarter (i) and area (j) the following equation was used,

(1)

where n is the number of respondents and y the reported harvest per respondent (k). The total population N is computed as:

(2)

where ρ is the number licenses issued being valid for a year (a), week (w) or day (d). The number fishing without a license (m) was computed using the estimated per-


centage that fished without a license even though obliged to have one (Table 3), mul-tiplied with the actual number of Danish citizens between age 18 and 65, which 1 January 2011 was 3,408,000 persons (Agerskov and Bisgaard, 2011). The values were weighted with the fishing effort ε which for those holding an annual license was de-rived from the omnibus survey and assumed to be 1 day for those holding a daily license and 3 days for those holding a weekly license. All values used can be found in Table 2.The standard error of (1) was computed according to Cochran (1977, page 37) as:

(3)

Where s’² is calculated as:

(4)

For (4) applies that Y´ij=Yij if the unit is in the ijth domain and 0 if not. The relative standard error (RSE) was computed as the standard error divided by the estimate. In the license list survey the respondent had the opportunity to report harvest in either kilo or counts. Therefore, it was necessary to find an average weight of a harvested fish in order to adjust from counts to kilo. The average size of eel and cod above minimum landing size caught in the passive fishery was found from Sparrevohn et al. (2009). Eel larger than the minimum landing size caught in fykenets was set to 47 cm corresponding to a weight of 188 g. Cod caught in fykenets above the minimum size was set to 39 cm corresponding to a weight of 540 g. Cod caught in gillenets was set to 47.5 cm, which corresponds to 975 g. Since no estimate on the average weight for cod caught and kept angling was available a value of 1,500 g per fish was chosen. For seatrout the average weight was set to 2,300 g when caught and kept in gillnets (Sparrevohn et al., 2009) and 1,700 g when caught and kept either in fykenets or while angling.

Table 2. Values used in eq. 1-4 for estimating harvest and RSE in passive gear fishing and an-gling. Effort is in days per year. The respondent number (n) given left to the slash is for the inter-view covering 1st and 2nd quarter and the value right is the interview covering the 3rd and 4th quarter.

Respondents (n)

Year Week Day Without Year Week Day Without (a) (w) (d) (m) (a) (w) (d) (m)

Passive gear

1904/1970 33,734 - - 16,609 30.8 - - 10.8

Angling 1789/1991 152,723 17,305 23,716 92,511 9.7 3 1 4.4

License (ρ) Effort (ε)


4.1.3 Results

4.1.3.1 Omnibus interview

During four interview rounds in October, November, December and January a total of 3,868 persons were interviewed. When asked whether they had fished within the last twelve months, between 13 and 16 % confirmed. Approximately 10 % of these were fishing with passive gear, 90 % were anglers and 0.1% fishing commercially.


Table 3. Table 3A shows the numbers of respondents (n) in the Omnibus in October-December 2009 and January 2010. In table 3B the numbers were scaled up to actual population size of person between 18 and 74 (N), which 1 Jan. 2010 was 3,416,369 (Agerskov and Bisgaard, 2010)

A Do you fish?Yes Yes No No- legal No-illegal %

Dec 968 Angling 116 58 58 30 28 24.1Passive gear 9 7 2 0 2 22.2

Nov 957 Angling 132 69 63 33 30 22.7Passive gear 17 8 9 2 7 41.2

Oct 958 Angling 119 59 60 34 26 21.8Passive gear 14 8 6 3 3 21.4

Jan 985 Angling 134 89 45 23 22 16.4Passive gear 21 11 10 3 7 33.3

B Do you fish?Yes Yes No No- legal No-illegal %

Dec 3,416,369 Angling 409,4 204,7 207,7 105,879 98,821 24.1Passive gear 31,764 24,705 7,059 0 7059 22.2

Nov 3,416,369 Angling 465,868 243,522 222,346 116,467 105,879 22.7Passive gear 59,998 28,234 31,764 7059 24,705 41.2

Oct 3,416,369 Angling 419,988 208,229 211,758 119,996 91,762 21.8Passive gear 49,41 28,234 21,176 10,588 10,588 21.4

Jan 3,416,369 Angling 472,927 314,108 158,819 81,174 77,645 16.4Passive gear 74,115 38,822 35,293 10,588 24,705 33.3

Do you have a license?(n)

Do you have a license?(N)


4.1.3.1.1 Fishing without license

For both groups of recreational fishermen a significant part was found not to have a license. However some did not hold a license due to a valid reason. Excluding the group that did not hold a license for valid reasons, 21% of all that claimed to have had angled within the last 12 months were doing so without a license (Table 3). For the passive gear fishermen, the number of people not holding a license is larger and on average for the four Omnibus surveys 30% fished without a license. The level fluc-tuated for passive gear fishermen between surveys properly due to fewer persons available in the latter group therefore interpretations from this data should be dealt with caution. Further there appeared to be a bias in separating between anglers and passive gear fishery in the first two omnibus surveys since some of the passive gear fishers gave meaningless answers to why they did not hold a license. For example, several passive gear respondents answered that they only fished in put & take, an answer that does not make any sense, since a fishery with gillnets or fykenets in put & take lakes does not exist. The problem was recognized and it was emphasized that respondents should have a clear understanding of the difference between anglers and passive gear fishery. In this investigation we have used the average for the four om-nibus surveys to up-scale the illegal fishery.

4.1.3.1.2 Effort

Since it was expected that that effort between fishermen holding a license and fi-shermen without was different, this was estimated in order to be account for in the total catch estimation (see eq. 2). Effort for those fishing without a valid license was investigated in the three latest omnibus where the respondents were asked about their fishing pattern and effort. Results indicate that for anglers fishing without a license, the effort was approximately one third compared to anglers fishing with li-cense. For passive gear fishers the effort for people without a license was a little lower than half compared to fishermen fishing with a license (Table 4).

Table 4. The average days fished for anglers and passive gear fishingermen that fished with ei-ther a license or illegally.

With license Illegally With license IllegallyNovember 09 8.5 2.2 24.0 10.7December 09 9.9 4.2 25.4 16.5

January 10 9.7 4.4 30.8 10.8

Angling Passive gear

4.1.3.1.3 Fishing in other countries

In the omnibus interview the respondents were asked about fishing habits abroad. The percentage that fished in other countries was in the four surveys between 2.1 and 3.3 % of all interviewed. Sweden and Norway were by far the most important coun-tries visited (Table 5). On average approximately 60 % reported one trip to other countries but some reported as many as 20.


Table 5. Respondents that fished in other countries from the four different omnibus surveys. Total numbers of respondents can be found in table 2.

October November December JanuarySweden 11 7 16 8Norway 9 3 6 5Faroe Island 3 2 0 2Great Britain 0 0 0 1Greenland 1 0 2 1Rest of Europe 4 3 7 1Rest of the world 3 6 5 7Respondents that fished outside of Denmark

28 20 32 25

4.1.3.2 License interview

The refusal rates were very low in the investigation as only 2% and 4% for the anglers and passive gear fishermen refused to answer (refuse and other reason, Table 6). The very high level of responses in the surveys qualifies the investigation.


Table 6. Distribution of non-respondents and there motive for not responding.

Anglers Total

1. Tlf 2. Web 9. No answer

1. Answer 1,726 2,054 0 3,782. Not meet 0 0 562 5623. Refuse 0 0 81 814. Other reasons 0 0 25 255. language problems 0 0 4 46. not relevant 0 0 0 07. No kontact on tlf.number 0 0 69 698. No tlf.number found 0 0 301 301Total 1,726 2,054 1,042 4,822

Passive gear Total

1. Tlf 2. Web 9. No answer

1. Answer 2,096 1,778 0 3,8742. Not meet 0 0 495 4943. Refuse 0 0 115 1154. Other reasons 0 0 44 445. language problems 0 0 5 56. not relevant 0 0 2 27. No kontact on tlf.number 0 0 39 398. No tlf.number found 0 0 245 245Total 2,096 1,778 1,02 4,894

Method

Method

We calculated the total estimated catches using 1) the total numbers of respondents, 2) only those that had answered vie the internet and 3) only those that had answered via the telephone (Table 7). A bias was clear when only using those answered via the internet as this underestimate the catches compared to the total estimates. Contrary, using only the internet answers overestimated the catches. This bias was connected to the fact that the percentage of respondents that had not fished was for both anglers and passive gear fishermen higher when answering via the telephone contra via the internet.


Table 7. Comparing the total estimated catch using either; all respondent, those that has answered via internet or those that answered via telephone. The percentage that had a valid license but has not fished in either 1st and 2nd quarter or 3th and 4th is also shown.

Angling All Phone InternetCod 1,164 t 803 t 1,463 tTrout 440 t 261 t 455 tHas not fished 1st and 2nd

41% 48% 36%

Has not fished 3rd and 4th

47% 59% 37%

Passive gear All Phone InternetEel in fykenet 116 t 90 t 145 tCod in gillnet 139 t 97 t 170 tTrout in gillnet 55t 37 t 62 tHas not fished 1st and 2nd

60% 63% 55%

Has not fished 3rd and 4th

48% 57% 40%

4.1.3.2.1 Passive gear fishers

A total of 4,894 persons were contacted and 3,874 volunteered to participate in an interview. 1,778 answered via the internet and 2,096 via the telephone survey (Table 6). Less than half of the passive gear fishers answered that they had been fishing within the last 6 months. In 2009 61 % answered that they had been fishing within the last 12 months. The respondents were asked to give their catches and fishing pattern on a three month interval. A total of 1,760 fished with gillnet, fykenet or a combina-tion (Table 8).

After completing questions about passive gear fishing and catches, the respondent was asked whether he/she also fished with rod. To that 63% confirmed but only 64% of those had actually fished within the last 6 months. This high level was also found in the 2009 interview which led us to analyze the fishery of this group separately from the rest of the anglers. A Fishery referred to as “angling with a passive gear license”.

Table 8. Different gear used in the passive gear fishing. Total represents all those that fish with either gillnet, fykenet or both.

TotalExclusively

gillnetExclusively

fykenetBoth

1,76 1,108 (62%) 261 (15%) 391 (22%)


4.1.3.2.2 Anglers

Of the 3,780 anglers that participated in the interview only 54 % had actually been fishing within the last 6 months, although they had a valid license.

4.1.3.3 Harvest

The total harvest estimate was upscaled with 20% for the angling fishing and 15% for the passive gear fishing due to the inclusion of the illegal fishing without a license.

4.1.3.3.1 Cod harvest

A total of 1666 t (RSE=7%) cod were harvested in the Danish recreational fishery in 2010 (Appendix A). Cod were harvested with all gears but with the main contribution (89%) from the angling fishery, 8% of the harvest derived from the gillnet fishery and only 3% came from fykenets.

The angling harvest of cod is quiet evenly distributed in the first three quarters of the year but in the last quarter only 13% of the harvest was taken.

The most important area for cod harvest were the Sound were 28% of the total cod harvest were taken followed by the Belt Sea with 25% and Skagerrak with 20%.

4.1.3.3.2 Eel harvest

A total of 116 t (RSE=10%) eel was harvested with fykenet in Danish recreational fish-ing (Appendix B). The majority (61 %) was taken during the third quarter. The most important area was the Belt Sea which alone accounted for 43 % of the total eel harv-est followed by the Sound with 10% of the total harvest.

4.1.3.3.3 Seatrout harvest

Seatrout was predominately harvested angling (90%) (Appendix C). In total 600 (RSE=6%) t was caught and kept annually and out of these 538, 2 (RSE=7%) was har-vested angling. For angling, the thee most important marine areas identified was the Belt Sea [157 t (RSE=10%)], Kattegat [91.6 (RSE=15%)] and the Arkona sea [72.3 (RSE=36%)]. The total harvest in freshwater accounted for 15 % of the recreational harvest.

4.1.3.4 Release

Opposite to the harvest the estimated release are provided in numbers since no aver-age weight of fish released are available. The numbers released per kilo harvested was calculated for each species on an area level (Fig. 2).


Figure 2. Number of fish released per kilo harvested.

4.1.3.4.1 Cod releases

It was estimated that more than 1,600,000 (RSE=7%) cod was released during 2010 in recreational fishing (Appendix D). The Belt Sea, Kattegat, Limfjorden and the Sound are those areas where the highest number are releases relative to the kilo harvested. Neither the average weight nor the survival of those cod released is known. The mor-tality of the cod released is likely to be high since the majority of the cod released [1,540,000(RSE=7%)] are released angling in quite shallow areas. Hence, the cod re-leased are unlikely to suffer from serious anatomical traumas caused by inflated swimblatter.

4.1.3.4.2 Eel releases

The number of eel released was the smallest of the three species investigated (Ap-pendix E). Around 50,000 (RSE=16%) individuals was released and the numbers per kilo harvested was less than 1, except for the Baltic were the estimated harvest and release was associated with very high relative standard errors (RSE=50% and RSE=75%, respectively). The mortality of the released eel is likely low, since all ell are caught with fykenets and eel in general is rather insensitive to physical disturbance.

4.1.3.4.3 Seatrout releases

More than734,000 (RSE=8%) seatrout was released in 2010 (Appendix F). The Arkona sea was the area where the lowest number of fish was released per kilo harvested. This corresponds well with anecdotic information that claims a high average size of seatrout in the area around Møn and Falster.


4.1.4 Discussion

In the present study the total Danish recreational seatrout, eel and cod harvest and release was found by; 1) estimating the harvest/release from a subsample of persons that has issued a license within the last 12 month and 2) estimating the amount of fishing carried out without a valid license. The latter was done using an interview survey targeting the entire Danish population between the age of 16 and 74, i.e. the omnibus survey. In the four omnibus surveys conducted the number of anglers be-tween 16 and 74 years which had practiced their hobby within the last 12 month was between 400,000 and 460,000. This corresponds to between 12 and 14 % of the Danish population which is very close to 12.5 % which was found in 1997 (Bohn and Roth, 1997). In another survey, relying on an interview panel, the number of anglers in Denmark was estimated to be 616,000 (COWI, 2010). The number of anglers that claimed to have issued a license was between 308,000 and 201,000 whereas the actual number of license issued - including weekly and daily licenses - is around196,000. According to the omnibus interview survey between 24,000 and 38,000 claimed they had a license for passive gear fishing which is close to the actual number of licenses sold, which is around 34,000. The margin between respondents that claimed to have a valid license and the actual number of license issued is relatively small. In 2010 the number of annual angler license issued were 152,723; weekly license was 17,305 and daily 23,716, summing to a total of 193,744, which is 20% lower than the persons that claimed to have had a valid license (an average value for the four omnibus survey).

4.1.4.1 Eel

In recreational fishing eels are mainly harvested in fykenets in saltwater, even though some freshwater fishing for eel exists. The intensity of the freshwater fishing is un-known since it can be carried out legally for all landowners along lakes and rivers. In the commercial fishery the landings from lakes are very low compared to those in saltwater. Of the total landings reported from 2005 to 2009 only between 2 and 3 % was from lakes (www.fd.dk).

Since fykenets set in saltwater are rather sensitive to wave and current action this fishing is mainly carried out in the inner Danish waters where wind and wave pro-tected Fjords, Belts and Sounds are located. This is reflected in the very low harvest of eel in the North Sea, Skagerrak and Eastern Baltic. The same pattern as last year was observed in 2010 where the Belts Sea was the area with the highest harvest followed by Kattegat and the Limfjord. Similar eel were not harvested equally throughout the season in 2010. In 2009 the highest harvest were reported in the period from August to October, which in 2010 roughly corresponds to the period from July to September. In 2009 the fishery with fykenets for eels was closed in the period from 10th of May to the 31st of July (Anon 2008). This is reflected in our survey as low harvest during the period from May to July in 2009 and April to June in 2010. Periods which tradition-ally has been months with a high CPUE of eel (Pedersen et al., 2005). The total har-vest, including fishery without license was in our investigation estimated to be 116 (RSE=10%) t which is a slight increase compared to the 104.4 (RSE=13%) t estimated in 2009 (Sparrevohn and Storr-Paulsen, in press). In 1997 the total catch of eel in the legal recreational fishery was estimated to be 138 t, which at that time corresponded to 20 % of the total eel yield (recreational harvest plus commercial landings) (Anon, 2008). The commercial landings were in 2010 on 408 t hence the recreational fishing harvested an equivalent of 22 % of the total Danish eel yield. In 2009 the figure was slightly lower at 19%.

http://www.fd.dk/


4.1.4.2 Cod

In 2009, Sparrevohn and Storr-Paulsen (in press) estimated that nearly 1,231 (RSE=6%) t cod were harvested in recreational fishing during 2009. In 2010 this was somewhat higher, i.e. 1,666 (RSE=7%) t. Similar to 2009 the main part of cod har-vested in the recreational fishery was angled.

Anecdotal information has highlighted the Sound as an important recreational cod fishing area but also The Belt Sea and Skagerrak showed in our survey high harvest. Commercial landing in the Sound has between 2004 and 2008 been fluctuated around 1,900 t (ICES 2010). The commercial landings in the Sound are mainly from a small area north of Helsingør called “Kilen” were it has been legal to trawl, opposite to the rest of the Sound where a trawling ban has been in place since 1932. However, a spa-tial and temporal closure (to protect the cod in the main spawning season) of the Sound commencing early in 2009 for both recreational and commercial fishery and reduced the commercial landings to 630 t in average the last two years (ICES, 2011). Due to the large decline in commercial catches later years the recreational fishing in 2010 accounts for 41% of the total Danish Sound cod yield and angling alone for 37 %. Angling harvest might be even higher, since cod harvest reported in numbers was converted into weight assuming an average mass of 1500 g. The average weight of cod caught and kept by anglers in the Sound is likely somewhat higher at least dur-ing the winter where spawning fish are targeted and fish larger than 10 kg are caught regularly. However, although the Sound was the area with the highest total recrea-tional harvest of cod it is not necessary reflecting an overfishing of the stock. Actu-ally, the Sound cod is considered to be in a relatively healthy condition, with a high CPUE and a wide age distribution compared to the adjacent waters (Svedäng et al., 2004; Svedäng et al., 2010). In the Western (SD 22-24) and Eastern (SD 25-32) Baltic Danish commercial fishing for cod accounted for 7,500 t and 10,700 t in 2010, respec-tively (ICES 2011). In this light recreational fishing was relatively important for the western area and minor in the eastern accounted for an equivalent of 12% in west and less than 1 % of the total cod yield in the eastern Baltic. Anecdotal information has highlighted a large fraction of German anglers fishing in the Danish part of the West-ern Baltic. However, it has not been possible to quantify the amount fished by for-eigners as it is possible in Denmark to purchase a license for a day or a week without providing any personal information. Therefore, it has not been possible to contact this fraction of the fishery.

In Kattegat, 66 (RSE=20%) t cod was harvested in recreational fishing; 53 (RSE=25%) t was from angling and 14 (RSE=21%) t from gillnet and fykenet fishery. However, due to the present very low commercial quota (270 t) and landings (117 t) in this area the recreational harvest are equivalent to 36% of the total cod yield in this area.

In the North Sea and Skagerrak the commercial Danish landings were by ICES esti-mated to be 5,700 t and 3,300 t, respectively in 2010 (ICES 2011b). The harvest in the recreational fishing from these areas was estimated to be 226 (RSE=24%) t and 330 (RSE=18%) t respectively corresponding to an equivalent of 4% and 9% of the total cod yield. Overall, our investigations indicate that a little increase in the total Danish recreational cod harvest happened in 2010 where 5.6% of the total yield was from the recreational fishery compared to 4.5 % in 2009 (Sparrevohn and Storr-Paulsen, in press).


4.1.4.3 Seatrout

For anglers in Denmark the most important species is seatrout, which it is caught both in marine and fresh waters (Rasmussen and Geertz-Hansen, 2001). A dispute between anglers and passive gear fishermen - about which of the two groups that is responsible for the main harvest - has been ongoing for several decades. This has resulted in e.g. restriction in the gillnet fishery where it has been prohibited to fish closer than 100 m from the coastline. This ongoing dispute could potentially influence the result of a recall survey. That 90 % of the harvest is likely to be taken by anglers and only 10 % by passive gear fishermen could therefore be biased in the sense that there could be a reluctance to admit high catches.

Like for cod in the sound, there might be a tendency for underestimating the harvest in the Arkona Sea, since this area, according to anecdotic information are an area with a high average weight. The same might very well be that case for fresh water where mainly mature individuals are caught. More than 734,000 (RSE=8%) are released every year by anglers. Besides undersized fish also spent and mature individuals are released. If we assume that the average weight of those fish harvested is 1,700 g then around 316,000 individuals are in harvested by anglers during 2010. Assuming that the average weight is correct then approximately every three out of four seatrout caught angling is released again. The number released in the passive gear was only 1.2 % of that released by anglers. This might reflect the fact that the mortality when caught in gillnets is probably much higher than when caught angling together with much higher size selectivity in gillnets.

4.1.5 Sources of error

Relying on respondent ability to remember catches or effort within a specific time period are followed by a number of biases such as digit preference, telescoping, non-responding bias and rule-based estimation. Digit preference means that the respon-dent will have a tendency for rounding figures to 0 or 5, a tendency that will increase with increasing recalling period (Huttenlocher et al., 1990). In this study we did see a tendency for some digit preference especially when reporting the catch in weight but whether this would increase or decrease the total estimated harvest is difficult to decide. Telescoping is the tendency for respondents to report an event, such as the catch of a trophy fish, even though it actually happened outside the time frame asked. This could potential mean an overestimation, especially in the angling harvest of cod, where some trophy fishing takes place. The bias introduced by non-respondents emerges since those fishermen with the lowest participation rate will have the highest non-responding rate (Tarrant and Manfredo, 1993), but since the non-respondent rate in present survey was very low this is not likely to have caused any major bias. Another potential source of bias is the risk that a rule is applied by the respondent when trying to remember the catches during the last 6 or 12 month. Typically, an average catch per trip is memorized and then multiplied with the as-sumed number of trips. This can potentially lead to a severe overestimation of the harvest, because there is a general tendency for exaggerating the participation rates in recreational events, there among fishing (Tarrant et al., 1993). For fishing it has been estimated that the effort was overestimated with 45% in a 12 month recall period compared to diaries (Connelly and Brown, 1995). This could impose a large overesti-mation in present study, especially for the passive gear fishing where it seems likely that some applies a rule, such as multiplying on average catch per gillnet or fykenet with the recalled number of days fished. This should be investigated further e.g. as suggested by the ICES Planning Group on Recreational Fisheries (ICES, 2010a) by a


dual frame approach where recall surveys are supported by either diaries or on-site surveys, such as access point interception or aerial based counting (Vølstad et al., 2006).

The angler recall survey only targets Danish citizens, even though tourist are also obliged to issue a license in order to fish legal in Denmark. In our study around 3 % of the Danes interviewed had fished as tourist in other countries, especially Sweden which is very close and easy accessible. There is no precise estimation of the number of tourist travelling to Denmark to fish, but the potential number of angling tourist is high. In Germany there is around 3,300,000 anglers (Anon., 2007) and for the Berlin-Brandenburg population around half claimed to have been on an angling holyday within the last year (Arlinghaus et al., 2008).

When estimating the harvest, the RSE for the passive gear estimates was in general higher than for angling. When computing the RSE it is assumed that the population sampled is infinite and if this assumption is not meet, then the RSE tends to be over-estimated. However, as long as the number of respondent does not exceed 5% of the population surveyed, the finite population correction can be ignored and the overes-timation will be minor (Cochran, 1977, p. 24). In the case of anglers less than 1% of the total number of anglers was included in the survey but for passive gear fishermen 4.6% of the population was actually sampled. Hence, it could be argued that – at least some of the elevated RSE – is caused by the estimation method. It could also be this group of fishermen actually is more heterogeneous than anglers. The heterogeneity of anglers and their behaviour patterns are unquestionable important and has been in-vestigated in several papers (Arlinghaus et al., 2008; Johnston et al., 2010), but whether these results can be applied to passive gear fishermen are not known.

4.1.6 Fishing without license

The inclusion of illegal fishing in was significant. Approximately 20-25 % reported that they fished illegally, though with a lower effort which corresponded to an in-crease in the passive gear catches on 17% and 24% for angling. One exception was in the November omnibus survey where 41 % of the passive gear fishers reported they fished without a license. However, there seemed - at least during the first interview round - to be a problem for respondents to differentiate between being fishing with passive gear (“fritidsfisker” in Danish) and angling (“lystfisker” in Danish). Indica-tion of some misunderstanding of the classification during the two first interview rounds in October and November was that respectively 3 and 2 respondents claimed that they did not need a license. As arguments for that they used reasons that do not make sense when fishing with a passive gear. E.g. claiming to only fish in put & take lakes. In December, where the confusion had been resolved none of the respondents claimed not to need a license. Therefore, this single high percentage of illegal fishery (41%) should be treated with caution. Another aspect when asking people whether they have fished illegally is the risk of under estimating the numbers since the re-spondents might be tempted to claim to hold a license when they actually do not.

4.1.7 Conclusion

Using a license list recall survey and including those fishermen that fished without a license showed that the recreational harvest was in some of the areas comparable to the commercial landings. This is a result of decreasing commercial landings more than it actually illustrates that recreational fishery in general imposes large fishing mortality. Nevertheless, it exemplifies that especially when stocks are overfished and


below its caring capacitive the fishing mortality caused by recreational fishing can be an important factor that should be incorporated into stock assessment, recovery plans and ecosystem bases management. The harvest of fishermen without a valid license was important as it increased the estimated harvests with 17% for the passive gear fishing and 24% for angling. Hence, recall surveys designed to estimate harvest and catches should not be based upon fishing license list alone but should also including those fishing without the mandatory license.

4.1.8 References

Anonymous, 2008. Danish Eel Management Plan. In accordance with COUNCIL REGULATION (EC) No 1100/2007 of 18 September 2007 establishing measures for the re-covery of the stock of European eel December 2008. © Ministry of Food, Agriculture and Fisheries, December 2008

Agerskov, U., and Bisgaard, M.P. 2010. Statistical Yearbook. www.dst.dk/yearbook, 2010.

Agerskov, U., and Bisgaard, M.P. 2011. Statistical Yearbook. www.dst.dk/yearbook, 2011.

Arlinghaus, R., Bork, M., and Fladung, E. 2008. Understanding the heterogeneity of recrea-tional anglers across an urban-rural gradient in a metropolitan area (Berlin, Germany), with implications for fisheries management. Fisheries Research 92: 53-62.

Bohn, J., E. Roth. 1997. Survey on angling in Denmark 1997 – Results and Comments. In: A.-L- Toivonen & P. Tuumaimem (eds) Socio-Economics of Recreational Fishery. Copenhagen: Nordic Council of Ministers, Temanord 1997, Vol. 604, pp. 79-88.

Cochran, W.G. 1977. Sampling Techniques, 3rd ed. Wiley, New York. 428 pp.Connelly, N. A., and Brown, T. L. 1995. Use of Diaries to Examine Biases Associated with 12-Month Recall on Mail Questionnaires. Transactions of the American Fisheries Society 124: 314-422.

COWI 2010. Analyse af adfærd, motive og præferencer blandt danske lystfiskere Udarbejdet som del af projektet Samfundsøkonomisk betydning af lystfiskeri i Danmark. http://www.fvm.dk/Admin/Public/DWSDownload.aspx?File=%2fFiles%2fFiler%2fFiskeri%2fsamf%26oslash%3bko+lyst+marts+2010%2fSamfundsoekonomisk_betydning_af_lystfiskeri_DK_version_final_12_03_10_samlet.pdf

Huttenlocher, D., Hedges, L. V., and Bradburn, N. M. 1990. Reports of elapsed time: bounding and rounding processes in estimation. Journal of Experimental Psychology: Learning, Memory and Cognition 16: 196-213.

ICES. 2009. Report of the Baltic Fisheries Assessment Working Group (WGBFAS), 22 – 28 April 2009, ICES Headquarters, Copenhagen. ICES CM 2009\ACOM:07. 626 pp.

ICES. 2010a. Report of the Working Group on the Assessment of Demersal Stocks in the North Sea and Skagerrak. ICES Document CM 2010/ACOM:13. 1048 pp.

ICES. 2011. Report of the Baltic Fisheries Assessment Working Group (WGBFAS), 12 – 19 April 2011, ICES Headquarters, Copenhagen. ICES CM 2011\ACOM:XX. In prep.

ICES. 2009b. Report of the Working Group on the Assessment of Demersal Stocks in the North Sea and Skagerrak - Combined Spring and Autumn (WGNSSK), 6 - 12 May 2009, ICES Headquarters, Copenhagen. 1028 pp.

Johnston, F. D., Arlinghaus, R., and Dieckmann. 2010. Diversity and complexity of angler be-havior drive socially optimal input and output regulations in a bioeconomic recreational-fisheries model. Canadian Journal of Fisheries and Aquatic Science. 67: 1507-1531.

Morales-Nin, B., Moranta, J., García, C., Tugores, M. P., Grau, A. M., Riera, F., and Cerdà, M. 2005. The recreational fishery off Majorca Island (western Mediterranean): some implica-tions for coastal resource management. ICES Journal of Marine Science 62: 727-739.

Pawson, M.G., Glenn, H., and Padda, G. 2008. The definition of marine recreational fishing in Europe. Marine Policy 32: 339-350.


Pedersen, S.A., J. Støttrup, C.R. Sparrevohn and H. Nicolajsen, 2005. Registreringer af fangster i indre danske farvande 2002, 2003 og 2004 – Slutrapport. DFU-Rapport nr. 155-05. 149s.

Rasmussen, G., P. Geertz-Hansen. 2001. Fisheries management in inland and coastal waters in Denmark from 1987 to 1999. Fisheries Management and Ecology. 8: 311-322.

Roth, E., A.L. Toivonen, S. Navrud, B. Bengtsson, G. Gudbergsson, P. Tuunainen, H. Appel-blad, G. Weissglas. 2001. Methological, conceptual and sampling practices in the surveying of recreational fisheries in the Nordic countries – experiences of a validation survey. Fish-eries Management and Ecology. 8: 355-367.

Sparrevohn, C.R., H. Nicolajsen, L. Kristensen, J.G. Støttrup 2009. Registrering af fangster i de danske kystområder med standardredskaber fra 2005-2007. Nøglefiskerrapporten 2005-2007. DTU Aqua-rapport nr. 205-2009. Charlottenlund. Institut for Akvatiske Ressourcer, Danmarks Tekniske Universitet, 72 p.

Sparrevohn, C. R.; M. Storr-Paulsen 2010. Eel and cod catches in Danish recreational fishing : Survey design and 2009 catches in series: DTU Aqua-report (ISSN: 1395-8216) (ISBN: 978-87-7481-110-7) , pages: 23, 2010, DTU Aqua, Charlottenlund

Sparrevohn, C. R.; M. Storr-Paulsen Accepted. Using interview based recall surveys to estimate cod Gadus morhua and eel Anguilla Anguilla harvest in Danish recreational fishing. ICES Journal of Marine Science. Accepted

Tarrant, M. A., and Manfredo, M. J. 1993. Digit preference, recall bias and nonresponse bias in self reports of angling participation. Leisure Sciences 15: 231-238.

Tarrant, M. A., Manfredo, M. J., Bayley, P. B., and Hess, R. 1993. Effects of recall bias and non-response bias on self-report estimates of angling participation. North America Journal of Fisheries Management 13: 217-222.

Toivonen, A.-L-., E. Roth, S.Navrud, G. Gudbergsson, H. Appelblad, B. Bengtsson, P. Tuuna-inen. 2004. The economic value of recreational fisheries in the Nordic countries. Fisheries Management and Ecology. 11: 1-14.

Vølstad, J.H, Pollock, K. H., and Richus, W. 2006. Comparing and combining effort and catch estimates from aerial-access designs, with applications to a large-scale angler survey in the Delaware River. North American Journal of Fisheries Management 26:727–741


Appendix A: Cod harvest (y) in tons per year with corresponding relative standard error (RSE). The number of respondents that reported a harvest within a given domain is denoted h.

y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE hJan- 0,9 87 2 0 98 1 2,6 67 4 3,5 100 7Apr-Jun 0,4 83 2 0 98 1 0 98 1 0 98 1 4,9 35 13 0,2 71 2 5,5 62 20

Jul-Sep 1,4 78 5 2 38 9 2,1 44 7 13,9 35 36 1,1 47 7 20,5 26 64

Oct-Dec 0 97 1 0,5 76 3 8 86 6 8,8 57 25 0,7 60 4 0 97 1 18 31 40

Total 0,4 83 2 1,4 78 7 3,4 34 15 10,1 69 15 30,2 25 78 2 34 13 0 97 1 47,5 17 131Jan- 2,6 70 2 1,1 62 3 0 98 1 2,1 56 6 6,4 65 11 9,2 29 29 6,4 48 9 5,6 54 5 33,4 18 66Apr- 0,9 64 4 4,5 59 7 0 98 1 2,5 33 16 2,4 36 11 5,8 23 49 4,3 47 14 8,6 55 9 29 21 111Jul-Sep 0,1 97 1 8,6 60 9 4,4 48 17 5,1 42 15 9,4 21 52 3,5 59 11 2,5 49 6 33,6 19 111

Oct- 0,5 86 2 8,4 59 7 0,2 97 1 1,2 53 7 17,1 92 11 11,9 27 37 2,3 68 8 1,3 69 2 42,9 16 75Total 4,1 48 9 22,6 34 26 0,2 97 3 10,2 26 46 31 53 48 36,3 13 167 16,5 27 42 18 32 22 138,9 9 363Jan- 5,9 44 9 11,6 36 14 0,1 98 1 1,6 60 5 16,7 23 43 16,3 28 44 5,1 46 8 2,9 65 7 60,2 12 131Apr- 16,1 23 28 20,9 21 38 1,5 98 1 9,7 56 12 19 30 41 22,6 23 60 24,6 75 18 6,1 39 15 120,5 17 213Jul- 13,7 72 9 23,8 29 25 0 97 1 3,6 45 15 21,5 22 50 15,8 30 50 3,8 56 10 8,6 64 8 90,8 16 168Oct-Dec 8,9 88 4 7,5 47 14 2,9 34 11 13 26 41 7,2 34 29 4,2 55 8 0,4 97 1 44,1 21 108

Total 44,6 30 50 63,8 15 91 1,6 92 3 17,8 33 43 70,2 13 175 61,9 14 183 37,7 50 44 18 35 31 315,6 9 620Jan- 55,4 57 8 39 70 7 5,2 82 3 110,4 19 73 80,6 35 42 4,2 59 3 5,5 60 5 311,9 16 141Apr-Jun 53,4 58 12 82,1 43 21 17,1 58 8 79,1 21 60 103,8 29 48 6,9 71 5 6,3 47 6 321,3 18 160

Jul- 52,5 48 11 112,9 34 21 0,8 77 2 9,7 43 10 93,4 16 79 80,9 32 51 58,8 74 12 9,5 70 7 385,7 18 193Oct-Dec 15,8 84 2 9,4 58 6 2,8 63 4 77,7 18 71 28,8 33 28 21,6 77 5 1,1 100 1 144,9 17 117

Total 177,1 30 33 243,4 24 55 0,8 77 2 34,8 34 25 360,6 9 283 294,1 17 169 91,5 51 25 22,4 36 19 1163,8 9 611Anglin 221,7 24 83 307,2 20 146 2,4 67 5 52,6 25 68 430,8 8 458 356 14 352 129,2 39 69 40,4 25 50 1479,4 7 1231Passive gear 4,1 48 9 23 33 28 1,6 69 10 13,6 21 61 41,1 43 63 66,5 13 245 18,5 25 55 18 32 23 186,4 8 494

Total 225,8 24 92 330,2 18 174 4 49 15 66,2 20 129 471,9 8 521 422,5 12 597 147,7 34 124 58,4 20 73 1665,8 7 1725

Grand total

Cod harvest angling with a

angling license

Cod harvest angling with a

passive gear

Cod harvest in

gillnets

Cod harvest in fykenets

Belt Sea Arkona Sea Eastern Baltic TotalLimfjorden KattegatCentral North Sea Skagerrak The Sound


Apendix B. Eel harvest (y) in tons per year with corresponding relative standard error (RSE). The number of respondents that reported a harvest within a given domain is denoted h.

y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h

Jan-Mar 0,1 98 1 0,1 98 1 0,7 90 3 0,3 54 4 0,8 36 12 0,3 67 3 2,3 77 24Apr-Jun 0 98 1 1,2 40 11 5,2 79 9 0,4 55 5 2,8 22 37 1 40 8 10,6 43 71Jul-Sep 0,6 52 5 6,6 89 2 11,3 19 50 11,9 19 58 6,5 62 15 28,1 17 99 4,9 24 26 1,3 57 5 71,2 12 260Oct-Dec 0,1 72 3 1,2 40 12 4,3 48 18 4,9 42 16 18,3 34 69 3,2 33 20 0,2 70 2 32,2 21 140

Total 0,8 42 9 6,6 89 3 13,8 16 74 22,1 23 88 12,1 37 40 50 16 217 9,4 17 57 1,5 50 7 116 10 495

Eel harvest in fykenets

Eastern Baltic TotalThe Sound Belt Sea Arkona SeaCentral North

SeaSkagerrak Limfjorden Kattegat


Appendix C. Seatrout harvest (y) in tons per year with corresponding relative standard error (RSE). The number of respondents that reported a release within a given domain is denoted h.

y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h

Jan-Mar 0 98 1 0 1Apr-Jun 0,1 98 1 0,1 98 1 0,8 88 3 0 98 1 1 72 6Jul-Sep 1 97 1 0,3 80 3 0 97 1 1,3 36 10 2,6 42 15Oct-Dec 2 97 1 0,1 58 3 0,4 75 2 0,4 44 6 0,1 97 1 3 99 13

Total 3 72 2 0,4 65 4 0,2 57 5 0,4 75 2 2,5 35 20 0,1 97 2 6,6 49 35Jan-Mar 0,1 98 1 0,1 98 1 0,8 65 4 1,7 33 10 0,4 52 5 1,7 35 13 1,2 57 4 6 33 38Apr-Jun 0,2 76 2 0,2 59 3 1,1 31 12 6 25 32 0,9 52 5 5,6 30 33 1,1 46 6 0,2 69 2 15,3 18 95Jul-Sep 0,3 58 3 0,8 97 1 3,5 31 21 7,6 26 34 2,8 37 11 6,1 20 52 1,3 54 7 0,3 97 1 22,7 13 130Oct-Dec 0,1 70 2 0,2 97 1 1,9 33 13 2,9 44 12 1,9 50 8 2,5 33 20 1,5 59 6 0,2 97 1 11,2 22 63

Total 0,7 37 8 1,3 63 6 7,3 19 50 18,2 16 88 6 25 29 15,9 15 118 5,1 28 23 0,7 54 4 55,2 8 326Jan-Mar 0,7 63 5 1,9 43 9 1,5 67 7 3,5 20 34 3 69 6 2,3 87 4 0,9 45 6 12,9 29 65

Apr-Jun 0,6 51 5 2,9 40 21 1,8 41 14 5,8 25 56 5,6 73 11 3 69 7 1,6 32 15 19,7 26 114

Jul-Sep 0,2 97 1 0,4 97 1 1,6 56 8 2,6 25 28 2,9 52 9 8 34 49 1 84 4 0,6 78 3 2,6 37 16 17,3 21 103

Oct-Dec 0,2 70 2 0,4 66 3 0,7 43 9 1 70 6 2 45 18 0,6 58 3 0,3 69 2 0,8 54 5 5,2 46 43

Total 0,4 60 3 0,4 97 1 3,3 33 21 8,1 20 67 7,2 29 36 19,3 17 157 10,2 46 24 6,2 47 16 5,9 21 42 61 11 367Jan-Mar 0,8 100 1 9,2 49 8 18,3 33 22 10,4 51 13 36,2 21 52 8,6 35 13 3,6 61 5 13,2 60 13 100,3 14 127Apr-Jun 2 60 3 0,2 100 1 10,1 41 14 25,6 30 38 16,1 41 23 45,7 21 81 6,5 40 11 4,8 37 10 22,1 26 26 133,1 11 207Jul-Sep 0,8 100 1 1,4 80 2 11,5 38 15 26,9 31 28 20,9 42 21 39,5 18 78 39 64 14 7,9 71 3 37,3 18 57 185,2 16 219Oct-Dec 0,8 100 1 3,4 38 9 12,7 40 21 6,6 46 9 16,3 28 37 8 40 12 2,3 54 4 8,5 30 18 58,6 14 111

Total 4,4 42 6 1,6 71 3 34,2 22 46 83,5 17 109 54 23 66 137,7 11 248 62,1 41 50 18,6 34 22 81,1 15 114 477,2 8 664

Angling 4,8 39 9 2 60 4 37,5 21 67 91,6 15 176 61,2 21 102 157 10 405 72,3 36 74 24,8 28 38 87 14 156 538,2 7 1031

Passive gear

0,7 37 8 4,3 54 8 7,7 18 54 18,4 15 93 6,4 24 31 18,4 13 138 5,2 27 25 0,7 54 4 61,8 9 361

Total 5,5 34 17 6,3 41 12 45,2 17 121 110 13 269 67,6 19 133 175,4 9 543 77,5 33 99 25,5 28 42 87 14 156 600 6 1392

Trout harvest angling with a

passive gear

Trout harvest angling with a

angling

Grand total

Belt Sea Fresh water Total

Trout harvest in fykenets

Trout harvest in

gillnets

Arkona Sea Eastern BalticThe SoundCentral

North SeaSkagerrak Limfjorden Kattegat


Appendix D: Cod release (y) in numbers per year with corresponding relative standard error (RSE). The number of respondents that reported a release within a given domain is denoted h.


Jan-Mar 1102 55 5 1102 55 5Apr-Jun 4159 98 1 21 98 1 104 80 2 2246 40 8 6530 64 12Jul-Sep 2191 90 3 583 71 3 1829 81 4 28338 35 19 201 97 1 33141 31 30Oct-Dec 201 97 1 4422 68 3 7778 44 14 804 71 2 13204 26 20

Total 4159 98 1 2412 82 5 687 61 5 6250 54 7 39463 27 46 1005 60 3 53977 21 67Jan-Mar 104 98 1 42 98 1 1165 48 7 1310 43 9Apr-Jun 208 98 1 21 98 1 166 62 3 458 89 2 2412 33 18 603 72 3 1664 68 3 5531 27 31Jul-Sep 101 97 1 161 62 3 1025 43 7 3758 25 23 241 83 2 382 62 3 5668 18 39Oct-Dec 101 97 1 60 97 1 3618 97 1 1467 47 9 80 97 1 80 97 1 5406 13 14

Total 409 60 3 21 98 1 492 38 8 5142 69 11 8802 17 57 925 52 6 2126 54 7 17915 11 93Jan-Mar 5895 44 9 10923 38 14 125 98 1 1618 59 5 109782 83 43 14165 27 44 5095 46 8 2932 64 7 150535 5 131Apr-Jun 16411 23 28 20661 21 38 1456 98 1 9934 55 12 20056 31 41 22244 23 60 23704 77 18 6195 39 15 120659 17 213Jul-Sep 13697 72 9 22654 30 25 20 97 1 3668 45 15 20003 22 50 79797 81 49 3819 56 10 8602 64 8 152259 43 167Oct-Dec 8944 88 4 7529 47 14 2968 34 11 13082 26 41 6672 36 29 4150 55 8 442 97 1 43787 21 108

Total 44946 30 50 61767 16 91 1601 89 3 18188 32 43 162923 56 175 122878 53 182 36767 51 44 18171 35 31 467240 15 619Jan-Mar 4803 47 8 4803 49 7 6845 59 4 85137 27 79 115277 38 46 1681 58 3 6724 65 6 225270 20 153Apr-Jun 6965 39 12 24376 53 22 12368 56 11 110834 32 64 103389 24 56 4443 69 6 12248 58 7 274623 11 178Jul-Sep 8848 56 11 78657 51 21 1079 100 2 21472 55 10 128937 30 84 143611 28 58 23629 36 14 7121 60 7 413354 14 207Oct-Dec 3561 72 2 8848 91 6 10466 61 5 55028 25 73 82110 27 33 6798 46 7 324 100 1 167133 15 127

Total 24176 27 33 116684 37 56 1079 100 2 51150 31 30 379935 16 300 444387 15 193 36551 26 30 26418 35 21 1080380 8 665Angling 69122 22 83 178451 25 147 2679 67 5 69338 24 73 542858 20 475 567265 17 375 73318 29 74 44588 25 52 1547620 7 1284Passive 4568 89 4 2433 81 6 1178 39 13 11392 43 18 48265 22 103 1930 40 9 2126 54 7 71892 16 160

Total 69122 22 83 183019 24 151 5113 52 11 70516 24 86 554250 20 493 615530 15 478 75247 28 83 46714 24 59 1619512 7 1444

Eastern Baltic TotalLimfjorden

Cod release in fykenets

Central North Skagerrak Kattegat

Cod release in gillnets

Cod release angling with

a angling license

Cod release angling with

a passive gear license

Grand total

The Sound Belt Sea Arko Sea


Appendix E: Eel release (y) in numbers per year with corresponding relative standard error (RSE). The number of respondents that reported a release within a given domain is denoted h.


Jan-Mar 62 98 1 104 98 1 83 98 1 250 41 3Apr-Jun 42 69 2 458 54 5 21 98 1 915 48 5 42 98 1 1477 34 14Jul-Sep 402 97 1 6029 97 1 7095 39 19 7034 30 22 2412 59 5 10431 27 27 1447 51 4 3417 75 2 38266 20 81Oct-Dec 402 97 1 603 68 4 2432 74 6 1889 93 3 4321 34 16 1005 79 3 20 97 1 10672 24 34

Total 804 69 2 6029 97 1 7739 36 25 9986 28 34 4426 51 10 15750 20 49 2494 43 8 3437 75 3 50664 16 132

Eel release in fykenets

Kattegat The Sound Belt Sea Arko SeaEastern Baltic

TotalCentral North Sea Skagerrak Limfjorden


Appendix F: Seatrout release (y) in numbers per year with corresponding relative standard error (RSE). The number of respondents that reported a harvest within a given domain is denoted h.

y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE h y RSE hJan-MarApr-Jun 42 98 1 416 98 1 458 90 2Jul-Sep 985 97 1 965 97 1 201 97 1 261 78 2 2412 57 5Oct-Dec 1005 97 1 201 97 1 302 65 3 1507 67 5

Total 1990 69 2 1006 93 2 402 69 2 979 51 6 4377 40 12Jan-Mar 104 98 1 104 98 1 208 65 3 104 98 1 520 38 6Apr-Jun 42 98 1 42 98 1 125 51 4 166 60 3 146 98 1 312 62 4 832 28 14Jul-Sep 201 97 1 221 58 5 884 44 11 40 69 2 1809 58 11 3155 36 30Oct-Dec 60 97 1 281 85 2 101 52 4 80 69 2 523 49 9

Total 303 69 3 42 98 1 627 44 11 1255 33 19 290 61 4 2409 45 20 104 98 1 5030 24 59Jan-Mar 458 98 1 624 59 6 2974 70 9 853 65 7 5490 25 40 478 46 7 250 60 5 940 45 6 12065 22 81Apr-Jun 333 98 1 1518 42 8 7236 61 23 624 54 15 5781 27 59 645 38 11 125 69 8 1649 32 15 17910 27 140Jul-Sep 0 1 1005 97 1 2633 43 8 3397 27 30 1487 65 10 9305 35 50 322 67 4 302 80 3 2593 37 3 21043 18 110Oct-Dec 241 70 2 402 97 1 563 64 4 804 65 10 1286 55 6 2291 37 20 523 78 3 60 73 2 752 54 4 6922 18 52

Total 1031 56 5 1407 75 2 5337 26 26 14410 35 72 4250 32 38 22867 17 169 1967 29 25 736 41 18 5934 21 28 57940 12 383Jan-Mar 961 100 1 9967 49 9 44189 35 26 10927 47 13 52475 28 59 9726 38 14 20894 86 6 9486 34 18 158625 18 146Apr-Jun 1561 72 3 240 100 1 9486 35 16 53796 62 48 15610 36 24 42388 20 97 4563 42 15 4923 64 11 12008 35 35 144576 24 250Jul-Sep 755 71 2 1834 94 2 30535 50 19 49848 32 34 25895 31 28 88799 18 98 27730 78 15 1187 91 2 41540 21 68 268124 13 268Oct-Dec 539 100 1 10790 46 11 11006 28 25 9495 59 14 37656 27 51 7445 73 14 2266 74 4 16616 38 25 95813 15 145

Total 3816 44 7 2074 84 3 60778 28 55 158839 157 21 61928 20 79 221319 11 305 49464 46 58 29270 63 23 79650 15 146 667139 9 697Angling 4848 36 12 3481 58 5 66115 26 81 173250 144 93 66178 19 117 244186 11 474 51431 44 83 30006 61 41 85584 14 174 725078 8 1080Passive 303 69 3 2031 67 3 1634 60 13 1657 30 21 290 61 4 3388 35 26 104 98 1 9407 23 71

Total 5151 34 15 5513 44 8 67748 25 94 174907 143 114 66468 19 121 247573 10 500 51535 44 84 30006 61 41 85584 14 174 734485 8 1151

Fresh water Total

Grand total

Limfjorden Kattegat The Sound

Trout release in fykenets

Trout release in gillnets

Trout release angling with a

passive gear license

Trout release angling with a

angling license

Central North Sea

Skagerrak Belt Sea Arkona Sea Eastern Baltic

ICES PGRFS REPORT 2011 | 39

4.2 England

4.2.1 New results in 2011 based on methods already described in the 2010 report.

None

4.2.2 New methods tested in 2011.

During 2008-09 the UK conducted a pilot survey of shore-based recreational sea an-gling (RSA) in Wales (Goodge et al. 2009, 2010) and reported this to the European Commission as part of the UK Data Collection Framework (DCF) National Pro-gramme. Previous surveys of sea angling in the UK have often focused mainly on economic value rather than making accurate estimates of catches. Building on the pilot study in Wales, and other European and worldwide surveys, a larger-scale sur-vey of recreational sea angling in England is being piloted in 2011 and 2012 to ad-dress the requirements of the DCF (Council Regulation (EC) 199/2008, Commission Regulation (EC) 665/2008 and Commission Decision 2010/93/EU) and the EU Control Regulation (Council Regulation (EC) No 1224/2009) for estimating specific recrea-tional catches. The surveys are also being designed to provide updated information on the economic value of sea angling in England.

A programme “Sea Angling 2012 – a survey of recreational sea angling activity and economic value in England” has been set up in 2011-12 for the following purposes:

• To provide information on RSA activity and economic value in England to allow the Government to make balanced and properly-informed decisions on management for sustainable development of all forms of sea fishing.

• To carry out surveys of RSA to meet EU legal requirements. • To provide sea angling bodies with useful information to allow then to de-

velop their own views and policies; • To provide the new Inshore Fishery Conservation Authorities (IFCAs) in

England with a project framework to help them develop policies on sea angling along their coasts, as well as demonstrating collaboration amongst the IFCAs and between IFCAs and stakeholders.

Sea Angling 2012 is a collaboration between the Department for Environment, Food and Rural Affairs (Defra), the Marine Management Organisation (MMO), Cefas, IF-CAs and recreational sea anglers. Stakeholder engagement is vital for delivering a successful survey, and a steering group is being set up including scientists, policy makers and members of the angling community. There will also be significant effort put into outreach to anglers in order to generate support for the project through arti-cles in the press, a website and through local tackle dealers.

This project is currently in the design phase, so finalised sampling schemes are not yet available. This is also a pilot project that aims to test a variety of methods for col-lection of fishing effort and catch per unit effort (catch & release) in order to develop estimates of total catch for certain species from shore, private boats and charter boats. The output from this work will be used to make recommendations for future cost-effective monitoring of RSA activity, and its economic value and impact. At this stage, estimation of catches from non-angling forms of recreational fishing (e.g. spear-fishing and netting) will not be included in the survey. It is expected that in the UK, such catches of the target species will be relatively small compared to angling catches.

40 | ICES PGRFS REPORT 2011

Four different surveys will be conducted, as described thereafter.

4.2.2.1 Inclusions of RSA questions in the Office for National Statistics (ONS) Opinions Module of the Integrated Household Survey:

The Opinions Module is a face-to-face interview carried out by the UK government at the resident’s house and responses are entered directly into a laptop. It uses random probability sampling broken into Government Office Region, proportion of house-holds with no car, socio-economic classification and the proportion of people aged over 65 years. The survey draws its sample from the Royal Mail's Postcode Address File (PAF) of 'small users' containing approximately 27 million private households in the UK. 2,010 addresses are sampled each month with probability of selection propor-tionate to size. The interviewer determines the household composition and then se-lects a respondent at random from those aged 16 and over. All household members over 15 are asked classificatory questions, but only one is selected to answer the Opinions Module questions and proxy responses are not allowed. The data are sub-sequently weighted to correct for the unequal probability of selection that this causes and are grossed up by age, sex and region to the population control totals used on the Labour Force Survey. As well as accounting for the unequal probability of selection, these weights correct for certain types of non-response bias and improve precision for most variables. We plan to include around 20 questions in the survey that allow the estimation of fishing effort, catch and economic value of RSA, including the location of angling activity. The focus will be on a recent period to reduce recall bias and will be repeated through the year to provide annual and seasonal estimates of activity levels. We will ask sea anglers to agree to take part in a further follow-up interview and the potential for a telephone diary programme will be assessed based on this response.

4.2.2.2 Charter Boat Survey:

Marine Conservation Authority documentation for head boats, charter vessels and for-hire companies is one source of listing angling boats in England, and further de-tails of the vessels can be complied by telephone (e.g. location, type and frequency of trips, species targeted, number of anglers, recent catches etc). Data are also available from recent vessel censuses carried out by Cefas contractors at English and Welsh ports to estimate numbers of vessels targeting bass. These two lists will be combined with local knowledge of MMO and IFCA staff to derive an operational register of recreational fishing boats that will form the sample target population. A telephone census of vessels on the operational register will be done to validate information held. A stratified direct interview survey involving completion of logbooks will be set up with and a record of the vessel activity requested over an agreed period. This will include the total number of trips in the survey period and a number of specific pa-rameters for each trip including: number of anglers on board; species targeted for the trip; type and location of activity (e.g. wreck fishing / indication of location of activ-ity); number and species caught; length and weight of individual fish; fishes caught; number of fish retained and returned; and estimate of number of fish surviving re-turn. Specific emphasis will be placed on the recording of species as required under the DCF and Control Regulation. Additional species will also be requested to validate the recording of information by respondents. Raising the catches to the overall fleet of charter vessels will be based on the area-season stratified vessel list, with appropriate adjustment for non-responses.


4.2.2.3 Shore & Private Boat Survey:

Information will be collected within this survey on fishing catch per unit effort, reten-tion, size composition and economic value. This will be done using two methods: an intercept / roving creel survey of individuals and a web-based survey. The question-naires for each of the surveys are still being produced and some the survey design is being developed. However, some details of the design of the intercept survey have been proposed. The population will be the total number of fishing trips conducted over a 12 month period and the sampling frame will comprise Primary Sampling Units (PSUs) representing coastal access points and days. The observer-based survey will be stratified spatially by ten IFCA regions and will coincide as far as possible with the periods being reported in the ONS household survey. Sampling intensity will comprise of around 360 one day sampling trips that will be selected using prior knowledge of angling intensity using random stratified or probability-proportional-to-size methods (using “expert” judgment of expected angling activity by site as measure of size). The probability of sampling at different times of day will be con-trolled using expert knowledge of diel fishing patterns in different areas and seasons. Information collected on the survey will include numbers of fish caught and released, duration of angling activity per individual, numbers of anglers on site, proportion of anglers not accessible by the ONS survey (e.g. overseas visitors) and (where possible) sizes of fish caught. Economic data will be collected from samples of anglers.

4.2.2.4 Economic Value and Impact Survey:

It is clear that economic value and impact is complex to measure and any comparison with commercial fishing is very difficult. A number of surveys have been done in the UK to measure economic value and impact (e.g. Drew 2004, Radford et al. 2009), but the most recent survey for England was Drew (2004). This needs updating to reflect current values, so a detailed survey that covers all forms of RSA and associated busi-nesses will be done within this project. Detailed survey design, data collection and analysis will be developed in collaboration with sub-contractors. This will utilise the methods in Drew (2004), Radford et al. (2009) and other worldwide surveys. It will include assessment of direct, indirect and induced effects, and investigate both eco-nomic value and impact.

The data from these surveys will be used to provide total catch and economic value. Total catches (retained and released separately) will be calculated using a number of different methods. For shore and private boats, effort will be estimated from the ONS survey and combined with CPUE estimates from the intercept survey (corrected where appropriate for avidity bias) to provide separate estimates of total catch. Com-parative estimates of CPUE from the ONS survey and the web-based survey will be obtained. Charter boat CPUE and catches will be estimated from logbooks on sam-pled vessels and raised to the total fleet of charter boats. Overall total catches will be obtained by summing over shore and boat anglers. Economic value and impact will be calculated from the data provided from all four surveys and, where possible, com-bined into single estimates.


References

Drew Associates. (2004) Research into the economic contribution of sea angling. Final report to the Department for Environment, Food and Rural Affairs, 71pp.

Goudge, H., Morris, E.S and Sharp, R. (2009) North Wales Recreational Sea Angler pilot sur-veys: Winter results December 2007 to March 2008. CCW Policy Research Report No. 08/14. 67pp.

Goudge, H., Morris, E.S and Sharp, R. (2010) North Wales Recreational Sea Angler (RSA) pilot surveys: Summer results July to October 2008. CCW Policy Research Report No. 08/31. 73 pp.

Radford, A., Riddington, G. and Gibson, H. (2009) Economic impact of recreational sea angling in Scotland. Technical Report to the Scottish Government. 259pp

4.3 Finland

4.3.1 New results in 2011 based on methods already described in the 2010 report.

The Finnish data collection strategy, with respect to the reliability, is to collect data on recreational fishing every second year. This practice will produce a continuous and methodologically stable biennial time-series of recreational catches. The new survey was conducted in the beginning of 2011 (referring to the year 2010). The new results for 2010 will be ready in autumn 2011.

4.3.2 New methods tested in 2011.

The methods used in Finnish recreational fishing surveys are described in the 2010 report. The methods were same in 2011 with two exceptions.

1. The inclusion probability of the sampling unit (household –dwelling) is the same for all sampling units inside a strata.

2. First time the questionnaire included the households fishing outside Finland. The number of fishermen was asked by country.


4.4 France

For more information on the general overview of the national recreational fisheries and on the detailed description on national recreational fisheries, see the 2010 PGRFS report.

4.4.1 Possible sampling frames

4.4.1.1 Species

4.4.1.1.1 Cod

No more information in 2011. See the 2010 PGRFS report.

4.4.1.1.2 Seabass

To evaluate the French recreational seabass catches, two methods were coupled: two telephone surveys and one diary database. The first telephone survey was conducted in 2006 at a national scale, for evaluating the overall French recreational fisheries characteristics. The second one was conducted in 2006 only in the coastal depart-ments of English Channel and Atlantic fronts and concerned particularly the seabass recreational fishery. The diaries were filled by fishers recruited during the second telephone survey, thus forming a panel. The telephone survey was used to estimate the number of recreational fishermen at a national scale and to increase the knowl-edge about their fishing practices. Diaries were used for collecting more accurate catch data, among which information on species, and fish lengths and weights.

An estimation of the coastal area production was estimated thanks to diaries, and coupling this database with the telephone survey information has allowed us to ex-trapolate the results and to obtain an estimation of the national seabass catches.

The telephone survey was conducted in both June and November 2009. The fishing logbooks from the panel’s fishers were collected during a whole year, until late 2010.

Sampling

A total of 15 090 households living along the French coasts of English Channel and Atlantic fronts was randomly sampled in June and November 2009. Among these ones, 258 fishers were volunteers to be part of a panel. They had to fulfil a fishing diary during a year. The collected information were:

fishing date gear used fishing travel duration (from their house to the fishing site and travel dura-

tion by boat if used) fishing place (town and fishing sector) port of departure fishing depth type of the fishing ground fishing duration seabass catches description

o weight o length o if kept or not

other catches description o species


o weight and number of kept fishes o weight and number of released fishes

These diaries were checked every 3 months during the 1-year survey.

A total of 1190 fishing trips were recorded, thanks to 190 fishers giving at least 1 fish-ing logbook during the survey. 68 fishers belonging to the panel did not fulfilled any logbook despite their volunteer action.

Effort results

Three coastal regions were particularly active through recreational fishing activities: Finistère, Morbihan and Manche (Fig. 3).

Figure 3. Coastal recreational activity targeting the seabass (from the panel database).

The seabass recreational fishing season is high between May and September, while this activity is continue during the whole year.

In July, 52% of the coastal fishers go fishing at least one time (Fig. 4). This value de-creases regularly in August (45%), September (39%)…until February (1%).


3% 1%6%

19%

40% 40%

52%

45%

39%

29%

17%

10%

0%

10%

20%

30%

40%

50%

60%

Jan. Feb. March April May Jun. Jul. Aug. Sept. Oct. Nov. Dec.

Figure 4. Proportion of fishers fishing at least 1 time during the month.

However, August was the month when fishers most go fishing, with a mean of 1,89 fishing trip this month per fisher (Fig. 5). During the cold season, meaning October to April, each fisher go fishing less than 1 time per month.

0,07 0,020,14

0,62

1,24 1,2

1,55

1,89

1,08

0,87

0,65

0,33

0

0,2

0,4

0,6

0,8

1

1,2

1,4

1,6

1,8

2


Figure 5. Mean number of fishing trip per month per fisher.

There is a mean of 0.81 fishing trip made per month per fisher, leading to a total of 2 177 378 fishing trips were made over the whole year by the coastal fishers. The mean duration of a fishing trip was evaluated at about 3.3 hours.

Around half of the panel fishers went fishing by boat, since 47% of them fished from the shore, while 53% fished from a boat.

Various fishing gears are used (Fig. 6): the fishing rod with lure represents 40% of used gears, followed by the fishing rod with living bait (23%).

Nets and spearfishing are less used, while nets represents 8% of the used gears and harpoon 4%.

Flyfishing is rarely used and represents around 0.2% of the fishing gears targeting the seabass.


2%

4%

4%

8%

20%

23%

40%

0,2%

Others

Flyfishing

Bottom line

Spearfishing

Net

Line

Fishing rod with living bait

Fishing rod with lure

Figure 6. Fishing gears used for targeting the seabass and their proportion

Catches

40% of the fishing trips were characterised by the catch of at least one seabass. How-ever, in half cases, a caught sea bass was released to the sea (Fig. 7).

Finally, the proportion of fishing trips leading to the catch and sampling of at least one seabass was of 30%.

During your fishing trip, did you caught at least a seabass?

Did you released a caught seabass to the sea?

Proportion of fishing trips with catches of at least one seabass, kept:

YES40%

NO60%

NO50%

YES50%

NO70%

YES30%

Figure 7. Proportion of seabass caught and kept by fishing trip.


All over the year, a mean of 1.2 seabass is caught per fishing trip and 0.6 is effectively kept (Fig. 8). However, we observed that in January, 1.7 seabass are caught per fish-ing trip, representing the highest value of the year. This also corresponds to the low-est value of kept fishes, with a mean of 0.3 kept fish per fishing trip per fisher.

0

0,2

0,4

0,6

0,8

1

1,2

1,4

1,6

1,8


CaughtKept

Figure 8. Mean number of seabass caught and kept during a fishing trip

A great scale of weights and lengths are observed in catches. The greatest proportion of seabass catches concerned individuals of about 500 g to 1 kg, representing 35% of the caught individuals (Fig. 9). This proportion decreases while the mean weight of the fish increases, and largest fish, heavier than 2 kg, represent only 8% of the catches. The kept seabass display an equivalent pattern. Only 10% of small seabass (under 500 g) are kept while around 30% of fishes weighting between 500 g and 1.5 kg are kept by fishers.


13%

15%

29%

32%

11%

7%

8%

16%

31%

38%

0% 5% 10% 15% 20% 25% 30% 35% 40%

> 2kg

1.5 - 2 kg

1 - 1.5 kg

500 g - 1 kg

< 500 g

CaughtKept

Figure 9. Proportion of caught and kept seabass by weight

The mean weight of caught seabass is estimated at 943 g and the mean length at 37.6 cm. The mean weight of a kept seabass is 1.253 kg and its length is of 46.3 cm.

Smallest fishes are caught in the greatest proportion (Fig. 10). 28% of the catches are composed of fish of less than 30 cm. However, only 4% of such small fishes are kept by fishers. The greatest proportion of fish kept by fishers are those measuring be-tween 36 to 42 cm (28%).


11%

24%

28%

27%

6%

4%

6%

13%

17%

24%

14%

27%

0% 10% 20% 30%

> 60 cm

50 - 59 cm

43 - 49 cm

36 - 42 cm

30 - 35 cm

< 30 cm

CaughtKept

Figure 10. Proportion of caught and kept seabass by length.

Finally, the total annual catches estimated for these coastal fisher, evaluated thanks to the panel database, is of 2 520 169 representing a total weight of 2 375.71 tons. How-ever, a great part of these fish are released to the sea and the kept proportion repre-sents 1 257 995 individuals, with a total estimated weight of 1 562.15 tons.

An extrapolation at the national scale was made thanks to the 2006 telephone survey database. An estimation of the fishers number in whole France, stratified by their fishing behaviour, was made. Thus, considering the more accurate data collected with the panel survey also stratified by fishing behaviour, it was possible to estimate the national seabass recreational production (Fig. 11).

It was estimated that 2 375.7 tons of seabass are caught by fishers from the coastal areas and 1 218.9 tons by fishes from the rest of France. Thus, a total of 3 594.6 tons of seabass is considered to be caught by all French recreational fishers.


DONT

2 079.6 t.

3 173.8 t.

1 546.5 t.

2 345.5 t.

=

=

CAUGHT KEPT

1 094.2 t. 799.0 t.

+ +

COASTAL

AREAS

OTHER AREAS

ALL FRANCE

Figure 11. Total production (in red) of the seabass (caught and kept fish) recreational fishery in France, stratified by area (coastal in blue and rest of France in green)

However, as a part of these fish are released at sea, the kept part is of 1 562.15 tons by coastal fishers and 805.7 tons by other fishers. Thus, the total seabass production in France is of 2 367.85 tons in 2009.

Socio-economic

We don’t have any socio-economic information at this time.

4.4.1.1.3 Bluefin tuna

Sampling

All recreational fishers in France have to report their bluefin tuna catches to the DPMA (Direction of Marine Fisheries and Aquaculture) since 2009. The preliminary results presented here concern the recreational fishing activity in 2010 in the Mediter-ranean Sea.

Effort results

In 2010 in France, a total of 616 recreational fishers have reported bluefin tuna catches.

Catches

The mean length of caught tunas is of 137.2 cm. The smallest one measured 115 cm and the longest one 247 cm. The mean weight of a caught bluefin tuna was of 44.17 kg. The smallest one was of 20 kg and the biggest one of 203 kg.

Based on the declarative database, is has been estimated that the total production of bluefin tunas caught by recreational fishers in the Mediterranean in 2010 was of about 27.21 tons.


4.4.1.1.4 Salmon and eel

No more information in 2011. See the 2010 PGRFS report.

4.5 Germany

4.5.1 General overview of national recreational fisheries

See the 2010 PRGFS report for the main description of the general overview of the German recreational fisheries data.

Eel is the main target species of leisure fishers. In 2011 a pilot study was initiated to estimate the landings of eel in the German marine coastal waters of the Baltic and North Sea obtained from these fisheries.

4.5.2 Detailed description of national recreational fisheries

See the 2010 PRGFS report for the description of the categories of recreational fishing, the geographic delineations, the water bodies, the platforms for fishing, the fishing gears used and the tournament fishing.

4.5.2.1 Management regulations affecting recreational fisheries

A permanent fishing license is required to fish in German waters. The fishing license is valid for a lifetime, once the fisher has passed an exam. In Schleswig-Holstein and Mecklenburg-Western Pomerania non-anglers can buy a limited “tourist license” without passing an exam. An annual duty has to be paid by active license holders in most German federal states. In Mecklenburg-Western Pomerania, an additional coastal fishing permit has to be acquired for recreational fishing in marine waters. This permit is not required for the recreational marine fisheries in Schleswig-Holstein (both Baltic and North Sea) as well as Niedersachsen (North Sea). Management regu-lations affecting recreational fishing are divers and vary between the federal states. The following table gives an overview of the existing management measures for the main species targeted in the North and Baltic Sea.


Table 9. Overview of the existing management measures for the main species targeted in the North and Baltic Sea.

Baltic Sea North SeaMain target

speciesManagement regulations

Management regulations

Cod MLS MLSEel MLS, S MLS

Flounder MLS -Garfish - -Herring - MLS

Mackerel - MLSPerch MLS n/aPike MLS, S, BL MLS, S

Pikeperch MLS, S, BL MLS, SPlaice MLS MLS, S

Salmon MLS, S, BL MLS, SSea bass n/a MLSSea trout MLS, S, BL MLS, S

Sole MLS MLSTurbot MLS, S MLS

MLS (= minimum landing size), S (= closed season), BL (= bag limit)

4.5.3 Sampling frames

See the 2010 PRGFS report for the recreational fisheries data.

4.5.4 Survey methods

See the 2010 PRGFS report for the recreational fisheries data.

4.5.5 Results

The following results apply to the Baltic Sea only, since earlier pilot studies revealed that there were no significant recreational fishery catches of cod in the North Sea. A pilot study for eel and shark is in preparation.

4.5.5.1 Sampling

To estimate the mean effort of anglers in 2010 (angling days/year) the results from the mail surveys 2004-2006 were used (pilot study). This data was augmented with the actual number of members in the angling associations in MV and SH, the number of fishery licenses sold in MV and SH and the annual numbers of angling licenses sold for the coastal waters of MV.

To estimate the catch per unit effort a total of 330 samples were realized in 2010. Thereby 59 samples were carried out targeting shore fishing activities interviewing 287 anglers and 271 samples were realized targeting boat and vessel angling yielding 2220 interviews.

The following table gives an overview of the sampling in 2010 to estimate the length composition of landings from beach fishing and boat/cutter angling.


Table 10. Overview of the sampling in 2010 to estimate the length composition of landings from beach fishing and boat/cutter angling.

Samples No. of measured cod (landed)

No. of measured cod (released)

Charter vessel trips with observer 55 296 2602

Boat- self-measurement 15 146 50

Trolling - self-measurement 9 80 13

Beach fishing – self-measurement 6 22 31

Shore fishing – fishing events 7 1042 -

Charter vessel – fishing events 10 1212 -

4.5.5.2 Effort

In 2010 a minimum of ca. 110,000 respectively ca. 155,000 anglers at maximum went fishing in the Baltic Sea. The total effort in the Baltic Sea in 2009 was estimated be-tween 938,595 and 1,614,490 angling days.

The following table provides an overview of the estimated effort for the different fishing types, bi-annual and in total (unit: angling days).

Table 11. Overview of the estimated effort for the different fishing types, bi-annual and in total (unit: angling days).

Type 1. Half year 2. Half year Total 1. Half year 2. Half year Total

Shore fishing 107,083 178,999 286,083 229,273 330,601 559,874

Wading 105,759 11,102 116,861 178,221 125,637 303,858

Boat fishing 111,936 167,356 279,291 251,675 293,313 544,987

Fishing charteredvessels

75,581 84,431 160,011 109,519 109,94 219,46

Trolling 7,805 8,233 16,038 12,426 12,088 24,514

Total 408,164 450,121 858,284 781,114 871,579 1,652,693

Minimum Maximum

The catch per unit effort (CPUE) – based on on-site surveys in 2010 – was calculated for the different fishing categories and was the highest for the boat/cutter fishing and the lowest for wading.


Table 12. CPUE calculated for each fishing categories.

Type Cod (landed) Cod (released)

Chartered vessel 3.3 2.4

Boat fishing 4.0 2.1

Trolling 2.9 0.9

Shore fishing 0.6 0.9

Wading 0.4 0.6

CPUE (catch/day) in numbers

4.5.5.3 Catches in numbers

The following table shows the cod catches (numbers) of the German recreational fish-ery in 2010, divided into released and landed cod, according to the applied fishing method, bi-annual and in total.

Table 13. Cod catches (numbers) of the German recreational fishery in 2010

Type 1. Half year 2. Half year Total 1. Half

year2. Half year Total

Shore fishing 26,831 267,649 294,481 70,118 490,987 561,105

Wading 67,686 7105 74,791 114,062 80,408 194,469Boat fishing 362,291 235,636 597,927 840,984 416,85 1,257,834Fishing cutters 106,368 155,573 261,941 190,85 228,269 419,119

Trolling 6501 9261 15,762 8984 13,969 22,953Total 569,677 675,224 1,244,902 1,224,998 1,230,483 2,455,481

Shore fishing 27,771 172,361 200,133 49,541 317,995 367,536

Wading 40,189 44,407 44,407 67,724 47,742 115,466Boat fishing 468,453 687,977 1,156,430 1,057,823 1,190,278 2,248,101Fishing cutters 299,951 430,7 730,651 380,174 510,389 890,562

Trolling 21,43 24,016 45,447 20,727 34,024 54,751Total 857,794 1,319,274 2,177,068 1,575,988 2,100,428 3,676,416

Released cod

Landed cod

Minimum Maximum

4.5.5.4 Catches in weight

In 2010 a minimum of 2059 t respectively 4264 t of cod at maximum were landed in the German recreational fisheries in the Baltic Sea.

The following table shows the cod catches of recreational fishing in tons in 2010, di-vided into released and landed cod, according to the applied fishing method, bi-annual and in total.


Table 14. Cod catches of recreational fishing in tons in 2010

Type 1. Half year

2. Half year Total 1. Half

year2. Half year Total

Shore fishing 7 108 116 19 199 218

Wading 18 3 21 31 33 64Boat fishing 99 95 194 229 169 398Fishing cutters 29 63 92 52 92 144

Trolling 2 4 6 2 6 8

Total 155 273 428 334 798 831

Shore fishing 16 98 114 28 325 353

Wading 23 2 25 38 49 87

Boat fishing 466 684 1149 1051 1629 2680Fishing cutters 298 428 726 378 699 1076

Trolling 21 24 45 21 47 67

Total 823 1236 2059 1516 2748 4264

Released cod

Landed cod

Minimum Maximum

Compared to the landings of the German commercial fishery in SD 22 + 24 in 2010 (4250 t), anglers landed 48% respectively 100 % of this quantity.

4.5.5.5 Fishing type

The majority of cod (85%) is landed by anglers fishing from private boats and angling cutters. The following diagram is based on landed cod catches in 2009.

Figure 12. Importance of fishing types based on landed cod catches in 2009


4.5.5.6 Seasonality/Trends

Based on the recreational fisheries survey data and the estimation method annual German cod catches in the Baltic Sea (SD 22 + 24) varied between 1907 t (2007) and 2766 t (2005) based on diary data. Using recall data annual catches varied between 2940 t (2007) and 4482 t (2005).

Table 15. Annual German cod catches in the Baltic Sea

Year Minimum Maximum

2004* 1959 3330

2005* 2766 4482

2006* 1934 3017

2007 1907 2940

2008 2152 3307

2009 2233 3387

2010 2059 4264

* Please note that yearly landings in the years 2004 - 2006 were calculated by means of average masses of the cod commercially caught in SD 22 + 24 and the length distribu-tion of landings of the anglers using a general length-mass relationship. Since 2007 calculations are based on the recorded length distribution of angler landings and the length-mass relationship from the German commercial fishery (from the active com-mercial fishery for boat, cutter and trolling & from the passive commercial fishery for surf fishing and wading).

Cod is fished year round but spatial variation of catches is high between the two German coastal states. In general, cod catches increase from west (SH) to east (MV) with the highest catch per unit of effort (CPUE) for retained fish on the isle of Rügen with on average 8.3 cod per fishing day in contrast to 0.8 cod caught around Flens-burg.

In addition to smaller catches, fish size decreases from east to west resulting in higher numbers of released cod in the coastal state of Schleswig Holstein (SH) with the high-est CPUE for released fish in Kappeln (SH) in June, namely 15.5 cod. The following diagrams depict the monthly CPUE of released (red) and retained (green) cod from cutter fishing in the two coastal states 2009.


released cod

landed cod


released cod

landed cod


released cod

landed cod


Next to spatial variations there are also temporal variations of cod catches. The fol-lowing diagram gives an overview of the length distribution of cod catches per quar-ter (cutter fishing, boat fishing and trolling).

released cod

landed cod


Figure 13. Overview of the length distribution of cod catches per quarter (cutter fishing, boat fishing and trolling).

4.5.5.7 Origin of anglers/tourism

During interviews anglers were asked where they reside. Only a minority of anglers came from the coastal states bordering the Baltic Sea. The majority of anglers came as tourist anglers from the further inland federal states. Very few foreign tourist anglers were encountered.

The following tables give an overview of the origin of anglers fishing from angling cutters in Mecklenburg Western Pomerania (MV).

Figure 14. Overview of the origin of anglers fishing from angling cutters in Mecklenburg Western Pomerania (MV).


The following tables give an overview of the origin of anglers fishing from angling cutters in Schleswig Holstein (SH).

Figure 15. Overview of the origin of anglers fishing from angling cutters in Schleswig Holstein (SH).

4.5.5.8 Data quality

As required by Council Regulation (EC) No 949/2008, data related to annual estimates must achieve precision level 1 (95 % confidence interval inside a deviation of plus/minus 40 %).

An analysis of the calculated landing data (2010) by means of bootstrapping esti-mated a relative deviation between 12 % as minimum and 21 % as maximum for the different estimated numbers of landings (see table below).

Table 16. Different estimated numbers of landings in Germany

Landings (numbers) 2.5 % Percentile 97.5 % Percentile

Diary effort-data,Jan - Jun 857,794 732,247 984,439

Diary effort-data,Jul - Dec 1,319,274 1,159,482 1,563,939

Recall effort-data,Jan - Jun 1,575,988 1,289,891 1,911,855

Recall effort-data,Jul - Dec 2,100,428 1,780,323 2,489,126

Confidence intervals (α = 0.025)


4.5.5.9 Primary customers for the data, and intended uses

Table 17. Primary customers for the data, and intended uses

European Commission

National government

Stock assessment

scientists

Academic researchers

Fishing industry

General public

1 Participation A, B, C A, B, C A, B, C A A

2 Fishing effort A, B, C A, B, C A, B, C A A3 Total catch

(retained/released) by species

A, B, C A, B, C A, B, C A A

4 Catch per unit effort by species

A, B, C A

5 Size/age distribution of

catchA, B, C A

6 Socio-economic data

N/A N/A N/A N/A N/A N/A

Cod, Cod, Cod, Cod,Eel Eel Eel Eel

Type

of d

ata

Key species (give list)

Cod

Customers for data

A: General monitoring of trends B: Stock assessment C: Monitoring of annual statistics relative to annual management targets for specific species D: In-season monitoring of cumulative statistics relative to annual management targets for specific species E: other (specify)

4.6 Latvia

In Latvia recreational fisheries could be divided in two basic parts: • Personal consumption fisheries operating by limited number of gears used

in commercial fisheries - gillnets and fykenets; • Angling operating by angling tackle (rods etc.).

Both these fisheries are regulated by different legislation. Recreational fishermen fishing with commercial gears are obliged to report the catches in the same way as commercial fishermen. These catches are included in total catch statistics.

Anglers are not obliged to report the catches, except salmon and sea trout in the riv-ers, were angling targeting these species is licensed. The share of angler’s catches in fisheries is estimated from inquiries, sold angling cards and returned licenses which contained information on catches.


Table 18. The share of recreational fisheries (t) in Latvia, 2009

Fisheries Angling Self consumption fisheries

Commercial fisheries

Recreational in % from total

Coastal 50 131 2513 6.7Inland 1500 67 260 85.8Total 1550 198 2773 38.7

Cod, salmon and eel are fish species which are required to be sampled according to Appendix IV of Commission Decision (2008/949/EC).

In DCP of 2009- 2010 Latvia obtained derogation for sampling cod.

In total the catches of salmon and eel by recreational fisheries in Latvia in 2009 esti-mated from two data sources:

• Log-books in self consumption fisheries, • Data from licensed angling, • Inquiry of anglers 2007.

The angling of salmon is allowed only in licensed angling in two rivers and the an-glers return information on catches. Thus information on catches of salmon in recrea-tional fisheries is exhaustive.

The catches of eel by anglers were estimated in inquiry carried out in 2007. The tar-geted angling of eel takes place mostly in few lakes where the eel has been artificially restocked. In several such lakes also licensed angling has been introduced and the catches of eel thus could be estimated. In other lakes and rives the catches of eel by anglers have occasional.

Table 19. The catch of cod, salmon and eel in (t) recreational fisheries in Latvia, 2009

Angling Self consumption fisheryCod 0 Not allowed 0Salmon 0.8 3.1 3.9Eel 1.2 0.2 1.4

SpeciesType of recreational fishery Total

recreational

Data from personal consumption fisheries are exhaustive because fishermen are obliged to report the catches by the same type logbooks as commercial ones. These data covers all seasons, gear allowed and watercourses.

Information on catches of salmon by anglers was obtained from reports on purchased and returned licences and is also exhaustive because only the licensed angling of salmon was allowed in 2009. Information on catches of eel by anglers was obtained from reports on purchased and returned licences and estimated by the inquiry.


2010 year data

Table 20. Share of recreational and commercial fisheries in coastal waters of Latvia, 20101

Species Commercial Self- consumptionTurbot 9.1 0.6Pearch 26.5 9.7Salmon 1.6 1.9Cod 95 Not allowedFlounder 173.8 36.6Herring 19346 20Sea trout 2.6 2.5Eel 0.7 0.2

1species included in LATVIAN NATIONAL PROGRAM FOR COLLECTION OF FISHERIES DATA

Table 21. The share of recreational fisheries (t) in Latvia, 2010

Fisheries Angling Self consumption fisheries


Recreational in % from total

Coastal 50 117 2411 6.5Inland 1500 75 204 88.5Total 1550 192 2615 40.0

Table 22. Number of fisherman’s and anglers in Latvia 2010

Fisheries Angling1 Self consumption fisheries


Coastal - 815 157Inland - 1216 273Total 100 000 2031 430

1estimated from the number of realised angler cards, 70000 in 2010, 5% from popula-tion are anglers

4.7 Netherlands

For more detail on the detailed description of the national recreational fisheries, on the possible sampling frames, on the available statistics, on the previous survey methods and on the pri-mary customers for the data, see the 2010 PGRFS report.

4.7.1 General overview of national recreational fisheries

The Recreational Fisheries Programme is managed and designed by IMARES Wagen-ingen UR in close co-operation with the Royal Dutch Angling Association (Sportvis-


serij Nederland). The current design of the Recreational Fisheries Survey not only collects information on catch and effort but also on motivation and satisfaction of recreational fishers and expenditure (lures, bait etc). Especially the last two items are of interest to the Royal Dutch Anglers Association to develop their strategies and policies to improve the quality of recreational fishing.

There are about 1.000 angling clubs in the Netherlands and eight regional federations which are a member of the Royal Dutch Angling Association (Sportvisserij Neder-land). Over 600.000 freshwater anglers are a member of this country-wide organiza-tion.

It is not allowed to use non-angling fishing gear (gill nets, fykes, long-lines) for rec-reational purposes in inland waters. From 2011 onwards, the use of fyke nets and long-lines by recreational fishers in marine waters has been banned. The future of the recreational gillnet fishery in coastal waters is currently under review by the Ministry of Economic Affairs, Agriculture and Innovation. The Dutch Recreational Fisheries Programme and this report will, until further notice, focus on angling recreational fishermen.

4.7.2 Recreational fisheries survey design

In the Netherlands, marine recreational fishers are not registered and are not required to obtain a recreational fishing permit. Therefore the most reliable survey should comprise of two components following Lyle et al. (2002) and Henry and Lyle (2003):

(1) Screening Survey: identify fishing households, profile fishing households, select participants for a follow-up, and

(2) Diary Survey: monitoring fishing (and economic) activity through regular contact (monthly) by survey interviewers.

Furthermore, a small-scale ‘on-site’ sampling program (pilot) has been implemented to provide additional independent data on catch, size and species composition of recreational fishers along the coast and charter boats.

In principle the programme will cover all types of recreational fishery in the Nether-lands and will collect data on all species caught in recreational fisheries in fresh and marine waters. For eel, also information will become available on the ratio of eel caught in marine and in fresh water. Screening Surveys (2009, 2012, etc) and 12 month Diary Surveys (2010, 2013 etc) are planned every three years. In 2011, priority will be given to the estimation of recreational catches of North Sea cod. In principle, new estimates of cod, eel and shark catches will be available in 2011, 2014 and so on.

Screening Survey

Biases in the Screening Survey were probably negligible due to careful design of the survey. The demographics of the frame population (56,730 households) is selected and maintained by one of the largest commercial marketing companies in the Nether-lands (TNS-NIPO) to ensure its frame population does not deviate from the demo-graphics of the whole Dutch population as determined by the Central Bureau of Statistics. The Screening Survey was offered ‘blind’ to the 56.730 households towards the end of December 2009. Every month the commercial marketing company (TNS-NIPO) sends a questionnaire about a range of topics (social, politics, products) to the households in its database. The households do not know what the topics are when they start filling in the online questionnaire and they are not allowed to skip topics or choose topics. The general (including questions on recreational fisheries) online sur-


vey of TNS-NIPO in December 2009 was completed by 45.518 households (109.264 people).

Diary Survey

During the Screening Survey, people were not only asked if they had participated in freshwater and/or marine recreational fisheries and if they wanted to participate in a 12 month Diary Survey but also to indicate roughly how often they had fished in the past 12 months to determine the level of fishing ‘avidity’ (1-5, 5-10, 10-25, 25-50, >50 annual fishing trips). As expected, the level of avidity was higher among the people that indicated to be willing to participate in the 12 month Diary Survey compared to the avidity of all the people in the screening survey. Knowing the “avidity” of log-book holders allowed us to correct the data to ensure the same avidity ratios when catches etc. were extrapolated to the whole population.

Participants of the Diary Survey were asked to maintain a logbook. Participants were approached on a monthly base by TNS-NIPO and requested to transfer the data re-corded in their logbooks to online questionnaires. Participants of the Diary Survey recorded per fishing trip detailed information on the fishing location, gear, catches (species, size), ratio kept-retained, reason released, motivation and satisfaction and expenditure.

Recall Survey

In addition, in April 2011 a Recall Survey, similar to the historical recall survey con-ducted by Sportvisserij Nederland, was conducted among 2000 recreational fisher-men to quantify the recall bias. Fishermen who indicated in the Screening Survey that they were willing to participate in the diary survey, but were not selected were asked, among others, about their catches and expenditure in the past 12 months (March 2010 – February 2011).

4.7.3 Preliminary results Recreational Fisheries Survey

4.7.3.1 Participation

Preliminary results show that at present there are 1.69 million recreational fishermen in the Netherlands, of which 1.05 million fish only in fresh water, 0.44 million fish in fresh and in salt water and 0.2 million fish only in salt water. Preliminary (uncor-rected) result of the 2011 Recall Survey showed that in 2010 50% of the recreational fishermen were not a member of a fishing club and 28% did not have a licence (VISPAS) or fishing permit. In 2010 650.000 freshwater anglers were registered with Sportvisserij Nederland. The Screening Survey indicated that 1.450.000 people par-ticipated in freshwater angling in 2010,it seems thus that more than 50% of the fresh-water anglers were not registered.


4.7.3.2 Seasonality

Figure 16. Seasonality in inland and marine recreational fisheries (data not corrected for Category and Avidity).

Preliminary results on seasonality in fishing trips and species compositions are shown in Figures 16 and 17. Clear seasonal patterns in fishing trips are observed in both inland and marine recreational fisheries, with low activity in winter and high activity in summer (Fig. 16).

Strong seasonality was observed in catches for several species in marine and inland waters. In the inland waters a marked increase of catches of pikeperch and pike was observed in winter. Seasonality in the catches of recreational fishers was more pro-nounced in the marine waters, with mackerel, garfish and sea bass regularly caught in summer while cod, whiting and dab were the common species in the winter months.


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100% OtherTenchPikeperchPikeRuddWhite breamCarpBreamRoachRedfin PerchEel

a)

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100% OtherSea bassWhitingSolePlaiceDabMackerelCodGarfishFlounderEel

b)

Figure 17. Species composition (release + retained) of the recreational fisheries in inland waters (a) and marine waters (b) based on preliminary results of the diary survey of 2000 fishers (data not corrected for Category and Avidity).

4.7.3.3 Target species

In order to save costs, the Netherlands used an online-diary system, whereby partici-pants were approached monthly for one year to record their catches online. Prelimi-nary results show that 55% of participants filled in their logbook all 12 months. Of the 45% that did not fill in their logbooks every month, most fishermen filled them in 9-11 times (Fig. 18). Follow-up surveys have been conducted to find out why fishermen did not fill their logbook every month so the number of completed logbooks is ex-pected to increase. It remains to be seen whether the logbook holders that dropped out of the program will affect the results. The avidity categories of the logbook hold-ers are known and can therefore be corrected according to the avidity categories of the recreational fishermen as determined in the Screening Survey.


Figure 18. Percentage of logbooks returned.

The preliminary estimates of recreational catches of eel, cod, sea bass and pikeperch are presented in Table 23. Overall, the recreation catch for all four species is rather significant (25-40%) in relation to the commercial catch. In comparison, the recrea-tional catch of sole is <1% of the commercial catch.

Table 23. Overview of recreational catches in the Netherlands in 2010 in tonnes. Note that these are preliminary estimates (uncorrected data; these data should not be reproduced without consul-tation with Martin de Graaf or Tessa van der Hammen IMARES).

Species Caught

(t) Retained

(t) Release

Rate (%) % Commercial

Catch Eel (Anguilla anguilla) 214 120 44% ~25% Cod (Gadus morhua) 840 795 5% ~25% Sea bass (Dicentrarchus labrax)

165 124 24% ~30%

Pikeperch (Stizostedion lucioperca)

993 168 83% ~30%*

*commercial cod fishery is limited by a quota (3219 t in 2010)

**expert judgement, assuming the same ratio in pikeperch catches between Lake IJsselmeer and other

inland waters as for eel catches.

Eel (Anguilla anguilla)

Rough estimates of annual recreational catches of eel fluctuate between 200 (Dekker et al. 2008) and 200-400 tonnes (Vriese et al., 2008). With the implementation of the Eel management plan, it is mandatory since 2009 for all recreational fishers in inland waters where the fishing rights are with the recreational fishers (clubs, federations etc) and marine waters (federal regulation), to release eel immediately upon capture. Given the ban to retain eel, the amount of landed eel is slightly surprising. However, the ban has only been in place for a short time while catching and eating eel by rec-reational fishermen has a long tradition. Clearly a considerable amount of campaign-


ing remains to be done by Sportvisserij Nederland to assure that eel is being released by recreational fishermen. On the other hand, the current landings are already con-siderably lower than previous estimates.

Cod (Gadus morhua)

In 2005 and 2006, a pilot survey was carried out under the Data Collection Regulation to estimate the catches of cod by anglers in the Netherlands (van Keeken et al., 2007). The annual catches were then estimated between 456-1765 tonnes and landings were estimated between 264-1037 tonnes. However, the estimates were considered very uncertain. An earlier study (Smit et al, 2004) estimated annual catches between 186-408 tonnes. In comparison, the commercial landings of cod in the Netherlands in 2005 and 2006 was about 2000 tonnes in each of those years. The different results in both studies can be mostly attributed to the different methodological approaches and assumptions. The current estimate is ~800 tonnes, which is about 40% of the commer-cial catch. Remarkable is also the low release rate (5%, table 23) compared to the study of van Keeken et al. (2007).

Sea bass (Dicentrarchus labrax) and Pikeperch (Stizostedion lucioperca)

It is not required by the Date Collection Framework to determine the recreational catches for sea bass and pikeperch. However, sea bass and pikeperch are both impor-tant species targeted by commercial and recreational fishermen.

A close look at the lengths of the fish recorded by the logbook holders seemed to show that many logbook holders might not actually measure the fish accurately. The lengths recorded were strongly biased towards 0s and 5s (e.g. 30, 35, 40 etc) and comparison with onsite data also suggests some overestimation of the sizes. Because length-weight relationships are used to estimate the weight of the catches, an overes-timation of the length also results in overestimation of the weights. Unfortunately the sparse amount of onsite data collected in 2010 were not sufficient to properly correct the logbook data. Discussion within the planning group and with Jeremy Lyle led to the conclusion that diary surveys are suitable to collect reliable data on the number of fish caught by logbook holders but are not suitable to collect accurate data on lengths or weights of fishes. Length or weight data will need to be obtained in well designed (spatially and temporally) onsite surveys. In Denmark similar unreliable length and weight data were observed in their surveys. In 2011 and 2012 an on-site pilot study should be initiated in the Netherlands to collect accurate length data of retained fish in the field.


4.7.4 Recall bias in historical catch estimates

0

100000

200000

300000

400000

500000

600000

700000

800000

900000

1000000

COD EEL SEA BASS PIKEPERCH

# la

nded

fish

DIARY SURVEY

RECALL SURVEY

Figure 19. Total number of landed cod, eel, sea bass and pikeperch caught in 2010 estimated with the Diary Survey and Recall Survey. Note that the Recall Survey overestimated the recreational catches with a factor 1.5 - 4.5. (uncorrected data)

In recent years several studies have been conducted on the catches and expenditure by recreational fishers. These studies were all based on phone and/or mail recall sur-veys. Due to the methods (recall surveys) the accuracy of these catch estimates are doubtful as recall surveys have been demonstrated to overestimate recreational catches by as much as a factor two. The inaccuracy of recall surveys was clearly dem-onstrated when the catches of recreational fishers estimated by the two different sur-vey designs are compared (Fig. 19). The recall survey overestimated the catches with a factor 1.5-4.5 in comparison with the estimated catches of cod, eel, sea bass and pikeperch with the diary survey.

4.7.5 Economic value Recreational Fisheries

Smit et al (2004) provided an overview of the contribution of the recreational fishery to the economy in the Netherlands. The expenditure of recreational fishers (men >15 years old) was determined by TNS-NIPO using an online questionnaire (recall sur-vey, 3816 households, 546 interviews with men >15 years old; Boutkan 2002). Accord-ing to the 2002 TNS-NIPO survey, the average male fisher spend €577 annually. This amount was relatively high compared to other (inter)national sources (NRIT 1988) on expenditure of recreational fishers and Smit et al. (2004) raised some concern about the methodology applied in the 2002 TNS-NIPO survey.

The difference between the estimates of expenditure by recreational fishermen ob-tained with the Diary Survey and the Recall Survey further illustrates that recall bias is a significant issue (Fig. ). Based on the Diary Survey the average annual expendi-ture of a recreational fishermen is €170 while the Recall Survey resulted in an esti-mated expenditure of €300. The contribution to the economy of the whole recreational fisheries amounted to €288 million.


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

LOGBOOK RECALL

feed

(artificial) bait

consumables (hooks, twine)

boat usage

maintenance durables

durables (rod, book)

permits

membership

insurance

food and drink

travel

accomodation

€ 170 €298

€ 288 M € 502 M

Figure 20. Expenditure of recreational fishermen as determined by the Diary Survey and the Re-call Survey. Note that the Recall Survey overestimated the expenditure with a factor ~2. (uncor-rected data)

4.7.6 Dutch recreational fishermen abroad

During PGRFS 2010 it became clear that estimating the amount of fishing tourist in a country was particularly hard to estimate because of the difficulty to intercept tour-ists. However, it is relatively easy to determine how many foreign fishing trips are conducted by their inhabitants, as part of their Screening and/or Diary Survey,.

A pilot-study was conducted in the Netherlands to estimate the number of foreign fishing trips by Dutch recreational fishermen. Participants of the Recall Survey in April 2011 were asked about their recreational fishing activities abroad (country, number of fishing days in inland water and marine water). According to the prelimi-nary results of the pilot study, Dutch recreational anglers spend 450.000 days fishing in foreign marine waters and 900.000 days fishing in foreign inland waters in 2010 (Fig. 5). Belgium, France and Germany were the most popular countries for fresh water fishermen. In the marine environment the most popular foreign destinations for recreational fishermen were Norway, Denmark and Croatia.


0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Other

Schotland

Greece

USA

Kroatia

Spain

Sweden

Austria

Italy

Norway

Canada

Denmark

Germany

France

Belgium

FRESHWATER

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Other

Dutch Caribbean

Sweden

Belgium

Germany

USA

Turkey

Italy

Ireland

Spain

France

Mauretania

Greece

Kroatia

Denmark

Norway

MARINE

(not corrected for Type of fishermen or Avidity)

450.000 fishing days 900.000 fishing days

245.000 fishing days125.000 fishing days

Figure 21. Preliminary results of the pilot study to estimate the number of foreign fishing days in marine water (left) and inland water (right) by recreational fishermen from the Netherlands.

4.7.7 Recommendations for the Netherlands

• Determine if the TNS-NIPO database is representative for the general population in the Netherlands. In December 2012, the Netherlands will have to conduct a screening survey using random digit dialling (phone) parallel to the online screening survey.

• Improve the onsite surveys. Conduct a pilot study in 2011-2012 to improve the onsite surveys to collect data on length of landed fish. One option would be to develop a program with Sportvisserij Nederland to recruit a group of “Volunteer Fisheries Liaison Officers” to assist in the collecting of data (catch, length, species composition) in the field.

• During the next Diary Survey in 2013 ensure that information on both fish-ing trips in the Netherlands as fishing trips abroad are recorded. An indi-cation of the number of fishing trips abroad may also be collected during the Screening Survey in December 2012.


References

Boutkan, A. 2002. Sportvisakte 2002; extra vragen. NIPO Rapport B-2730. Amsterdam.

Dekker, W., C. Deerenberg & H. Jansen. 2008. Duurzaam beheer van de aal in Nederland: onderbouwing van een beheersplan. Wageningen IMARES Rapport C041/08. Pp. 99.

Henry GW, Lyle JM (2003) The national recreational and indigenous fishing survey. FRDC Project No. 99/158. NSW Fisheries Final Report Series No. 48, pp 188.

Lyle JM, Coleman APM, West L, Campbell D, Henry GW (2002) New large-scale survey meth-ods for evaluating sport fisheries. In: Recreational fisheries: ecological, economic and so-cial evaluation, TJ Pitcher, C Hollingworth (eds), pp 207-226. Blackwell Science.

NRIT. De Economische betekenis van watersport en sportvisserij in Nederland. Nederlands Research Institute voor Recreatie en Toerisme, Breda, 1988.

Smit M, de Vos B, de Wilde JW. 2004. De economische betekenis van de sportvisserij in Nederland. Den Haag, LEI, Rapport 2.04.05, 75p.

Van Keeken, O., A. Dijkman Dulkes & P. Groot. 2007. Pilot study: Catches of North Sea cod by recreational fishermen in the Netherlands. CVO report 07.002. Pp. 28.

Vriese, F.T., J. Klein Breteler (VIVION), M.J. Kroes & I.L.Y. Spierts. 2007. Beheer van de aal in Nederland. Bouwstenen voor een beheerplan. Visadvies Rapport VA2007_01. Pp. 174.


4.8 Norway

No new results are available for 2011. See the 2010 PRGFS report for the Norwegian recrea-tional fisheries data.

4.9 Poland

No new results are available for 2011. See the 2010 PRGFS report for the Polish recreational fisheries data.

4.10 Spain – Basque country (Atlantic area)

No new results are available for 2011. See the 2010 PRGFS report for the Basque country recreational fisheries data.

4.11 Spain – Balearics Islands (NW Mediterranean)

See the PGRFS 2010 report for more information on the general overview of the Spanish Mediterranean and Balearic islands recreational fisheries, the detailed description of the na-tional recreational fisheries, the possible sampling frames, the available statistics, the survey methods and the primary customers for the data.

4.11.1 New results in 2011 based on methods already described in the 2010 report

4.11.1.1 Socio-economic profile of the recreational fishermen in the Balearic Islands

Data from the fishing licenses

It is compulsory to have a fishing license since 2000, there are several modalities reaching to 50,003 operative licenses for the Balearic Islands in 2009, distributed as follows:

• Individual fishing license valid for 2 years: N=43328 with fishermen aver-age age of 45.81 yrs.

• Boat license for boats registered in the 7th list up to 5 anglers on board and valid for 3 years: N=5002 with an fishermen average age of 48.83 yrs.

• Submarine individual fishing license valid 1 year: N=1673 with a fishermen average age of 34.71 yrs.

• Collective boat license from the 6th register list for up to 12 anglers on board valid 1 year with a licensed fishing master on board N=16.

Since the beginning of the licensing system the number of licenses has increased, although the tendency is not significant when compared between years or Islands. The practitioner’s average age is relatively stable for the period 2002-2009.

Mallorca Island with most of the Balearic Islands population represents 70% of the total fishing licenses issued. The number of licenses per 1000 inhabitants is 39.3, lower than in Ibiza-Formentera 58.34 or Menorca 61.6 per 1000 inhabitants.

Results from a mail survey

The Mallorcan recreational fishermen that marked in the form for the fishing license that they were willing to cooperate in a study, were contacted by mail as described in 2010 report. The method was a variant of the tailored design method (Dillman, 2000).


The response rate was 46.2% from the 734 forms send. The main issues limiting the number of potential answers were that more than one license was issued in the same address and errors in the addresses registered in the data bank (18%).

Social profile

Most RF have a single license modality (see 1.1) (74.34%) with 19.7% having 2 modali-ties. Albeit they practise at least 2 fishing modalities, being the choice very wide-spread. The boat fishing is significant (43.07%) albeit the majority fish from the coast (56.93%).

The social profile of the recreative fishermen (RF) was a male (91.15%) with an aver-age age of 44.35 ±15.59 years old, with a wide range of professional activities, albeit with a high proportion of retired people (19.47%). The average experience in this leisure activity was 22±16 years. Only a small proportion belonged to a fishing club (14.16%).

The education level of the RF was mainly basic (38.05%) with equal proportions of universitary studies (21.53%) and technical levels (21.53%). The income of the RF was comprised predominantly between the 800-1500 €/month (32.45%) and 1500-2500 €/month (28.91%) salaries.

Regarding the size of their household most RF belonged to households with 2 (26.84%), 3 (23.6%) or 4 (28.61) members. The majority of RF (60.77%) had a family member that was an active RF. This relates to the fact that the majority had learned to fish with a family member (52.8%) or with friends (31.56%).

Motivations

The main motivation for practising this traditional activity was for pleasure (92%), consuming the captures was less important (34.2%) and to follow a family tradition the less relevant (14.2%) to choose this leisure activity.

The fishing zone was selected based on the tranquillity (50.2%) and the captures ob-tained (39.5%). Secondary motivations were the proximity and access ease (26% each).

Regarding the use of the captures the main was fresh consume (72%), albeit depend-ing of the size of the capture part was frozen for late use (40.1%). A small part of the capture was released (20.1%).

Economic information

The data gathered from the enquiries allowed to differentiate the expenditure in the general practice of the activity from the expenditure of the RF fishing from a boat and linked to the boat.


Table 24. The annual general individual expenditure (excluding the license costs) was as follows based on 306 answers

CONCEPT MEAN € S.D.Fungible 181.23 279.7Reels, hams, etc 131.08 223.64Others 45.72 105.11Championships’ participation 0.78 5.9Lodging 8.12 52.47Food 22.16 43.44Transport 18.13 41.56Annual Mean 427.82 523.03

The expenditure linked to the boat ownership was very high reaching 3100 €/year mainly due to maintenance costs. The observed variability in the running expendi-ture was also found in the boat expenditure with a s.d. of 4404 €/year. This high vari-ability in the costs required to deal with the data in a special way to define better the general economic trends.

Table 25. Expenditure for boat owners.

CONCEPT N Mean S.D.Berth 123 735.20 1148.35Trailer 123 38.82 76.64Launch 123 40.09 81.64Formation 123 29.80 167.15Harbour taxes 123 46.98 143.26Transport 123 67.48 149.63Storage 123 129.49 517.94Mantenance 123 566.95 839.33Insurance 123 217.66 303.71Fuel 123 641.06 1470.11Electronics 123 310.81 1687.88Tournements 123 42.44 450.81Friends (bevrage, food) 123 233.9 1631.84Total 123 3100.69 4404.97


4.11.1.2 Valuation of the recreational fishing practised by tourists in Mallorca

During August 2009, 1,504 face-to-face surveys were conducted in Palma´s Port (440) and Airport (1,064), the two main points of entry for tourists in the island. Surveys were conducted from Monday to Friday during four days in Palma´s Airport, while the passengers were at the boarding gates waiting to take their plane, and six days in Palma´s Port, where the passengers take their boarding cards. The survey was fo-cused on tourists who were finishing their holidays in Mallorca. The languages used during the survey were Spanish and English.

Pollsters asked randomly to one person from each group or family. If the interviewed was resident in Mallorca was not considered. The first question was if the tourist had been fishing during his/her holidays in Mallorca. For those who said “yes” the poll-ster requested them to answer 14 questions regarding their profile and their fishing experience in Mallorca. For those who said they hadn´t been fishing in Mallorca the pollster asked them if they used to fish in their habitual residence:

Figure 22. Description of the survey process

Once the survey data were analyzed the results were extrapolated to the entire tourist population, based on data offered by the Balearic Government in its annual Tourism reports. To calculate the average catch and the average expenditures it was consid-ered one interview as one non-resident recreational fisher (NRRF) and each NRRF represent one trip to Mallorca. Each trip to Mallorca includes all the fishing outings (or fishing day) the NRRF did. One polled tourist means, in this way, one fishing experience in Mallorca.


The response rate was 96.3% (Airport: 96.7%, Port: 95.2%), resulting in 1448 complete interviews corresponding to 71.1% (N=1,029) in the airport and 28.9% (N=419) at the Port. The majority of the surveys were answered by men 63.1% (N=914) and the other 36.9% (N=534) by women.

A Pearson Chi-square test did not show differences in the proportion of NRRF, po-tential NRRF or no NRRF between the sampled population at the Port and the Air-port (χ2 = 2.576; df = 2; p = 0.276), therefore the results of the surveys are pooled together irrespective of the arrival way. From the 1,448 interviews performed, 2.35% (34 of those polled) said they had been fishing in Mallorca (NRRF) and 9.25% (N=134) said they had not been fishing in Mallorca, although they used to fish in their home place (potential NRRF).

Table 26. Results of the questionnaires.

Form 1 Form 2Number of surveys 134 34Percentage (%) 2.35 9.25Response rate (%) 99.2 88.2Gender (men) (%) 84.3 73.5Under 31 23.3 43.331-50 51.9 33.4Over 50 24.8 23.3Spanish 39 (29.3%) 15 (50%)Foreigners 94 (70.7%) 15 (50%)Number of nationalities 19 9Sure yes 15.8 73.3Probably yes 44.4 13.3Don´t know 9 3.3Probably not 8.3 10Sure no 22.5 0

Age (%)

Willingness to fish in case they

came back to Mallorca (%)

The 9.25 % of the total interviewed were non-resident recreational fishers who did not fish during their Mallorca holidays.

Answers to the question “why haven´t been you fishing in Mallorca?” are in Table 27. The 35.7% of respondents founding a lack of information pointed out a lack in the information regarding legislation, the 35.7% regarding the main places to practice recreational fishing and the remaining 28.6% pointed out a widespread lack of infor-mation.

There were differences between the two groups – Spanish and Foreigners – in their responses to the question “Why haven´t been you fishing in Mallorca during your holidays” (χ2 = 8.063; df = 3; p = 0.04473). The motivation “Spend their time enjoying other activities” includes meteorological reasons (3%) and the absence of the recrea-tional fishing license required (1.5%). There weren’t any differences between the two groups in the willingness to fish if the respondents came back to Mallorca (χ2 = 5.382; df = 4; p = 0.2503).


Table 27. Answers to the question “Why haven´t been you fishing in Mallorca?”

Reason Percentage (% of the total answers)

% over the total Spanish number

% over the total Foreigner number

Spend their time enjoying other activities 69.2 59 73.4

Don´t have their fishing equipment 15.8 23.1 12.8

Lack of information 10.5 17.9 7.7Lack of offer / facilities 4.5 0 6.4 The 2.35% of the total interviewed people said they had been fishing during their stay in Mallorca. There was similar age composition between the non-resident recreational fishers who fished in Mallorca and those who did not (χ2 = 10.956; df = 6; p = 0.08974). Only 3.3% (N=1) of those who fished during their holidays in Mallorca said the pos-sibility to practice recreational fishing played an important role to choose Mallorca as his holiday destination. Recreational fishing role was mainly “insignificant” (63.3%, N=19) or “not important” (33.3%, N=10) for those tourist who were fishing in Mal-lorca.

Regarding their fishing experience in Mallorca 40% of the NRRF (N=12) were consid-ered as “habitual NRRF” (it means they had been fishing at least one time during the last year in other locations) while the remaining 60% (N=18) stated don´t used to practice recreational fishing although they did in Mallorca during their holidays be-ing considered “occasional NRRF”. Most of the NRRF were satisfied (43.3%) or highly satisfied (33.3%) with their fishing experience. There were no differences in the fishing experience or satisfaction regarding the fishing activity between “Foreigners” and “Spanish” (χ2 = 0.02437, df = 1, p = 0.876 and χ2 = 5.959; df = 4; p = 0.2022).

The average stay in Mallorca for those tourists who practiced recreational fishing was 16 ± 8 days and the average of fishing days was 4 ± 2.7 days. It represented about the 30% with regard to the whole days they spend in Mallorca. The average of hours per fishing outing was 2.6 ± 1.3 h. The fishing modalities used by NRRF were angling from shore (80%), angling from boat (20%), trolling line (10%) and spearfishing (7%).

The total catch and expenditures declared by NRRF (N=30) were 48.5 Kg and 926 € respectively. The resulting average catch and average expenditures was 1.6 ± 2.7 Kg / NRRF and 30.9 ± 91.3 € / NRRF respectively. “Habitual NRRF” declared an average catch of 0.8 Kg / NRRF while “occasional NRRF” declared an average catch of 1.8 kg / NRRF respectively.

Recreational fishing expenditures are the total amount NRRF spend in bait, tackles (rented or bought) and fuel to arrive to the fishing zone is considered. Bait repre-sented the 15.9%, fuel the 11.3%, purchased fishing equipment the 15.6% and rented fishing equipment the remaining 57.2% of the total expenditures.

Both, Spanish and Foreigners NRRF showed similar trends in terms of duration of the trip in Mallorca, fishing days, total catch declared and total expenditures related to the recreational fishing activity (ANOVA test). Moreover, there were no differences between the two groups in terms of fishing modalities (χ2 = 0; df = 1; p = 1) and the possibility of fish again in case they came back to Mallorca (χ2 = 5.727; df = 3; p = 0.1257).

Considering the 2.35% NRRF found and taking into account the 8,718,788 tourists who arrived to Mallorca in 2009 (INESTUR-CITTIB, 2009), the NRRF number in Mal-lorca in 2009 was estimated at 204,892 with a confidence interval of 2.58% (Cochran,


1977). Considering the confidence interval the total number remains between 0 and 429,836 NRRF. Considering that 9.25% stated that they used to fish in their country despite they did not fish in Mallorca and extrapolating to the total number of tourists arrived in 2009 the potential NRRF number is 806,487. Taking into account the confi-dence interval the number remains between 581,543 and 1,031,433 potential NRRF.

This highlights the importance of the tourism in the recreational fisheries in Mallorca Island providing figures to an activity which existence was supposed but not quanti-fied. The recreational fishing as a potential economic input has not been seriously considered in the Balearics neither as a possibility to diversify the touristic offer. De-spite the huge number of NRRF and potential NRRF Mallorca received in 2009 they seem to be interested in recreational fishing as an extra leisure activity more than a high extractive activity in which case recreational fishing could be developed as a tourism supply involving both recreational and commercial fishers economic sector. If managers consider the improvement of this activity as a complementary offer, one of the main points would require changes in the licenses allowing for temporal per-mits and improvement of the information system. Moreover, the offer of charter boats and other modalities should be improved. For instance, commercial fishermen could be involved in the tourist recreational fishing business if changes in the Balearic fish-eries legislation support recreational fishing as an Ecotourism activity. More work is needed to better define the tourist recreational fishers profile and their likes.

4.11.1.3 Nautical tourism

As reported in 2010 on board enquiries were performed in summer 2009 using our own boat. The sampling was performed during 11 days covering a sizeable part of the coast. The nautical tourists considered include the Balearic Islands residents that were not in their port base following the official definition. A total of 422 boats were visited and 96% answered the questionnaire (N=405).

Figure 23. Sampling area (blue dots) in the different coastal areas off Mallorca Island.

The 80% of the respondents were male, mainly between 41-50 yrs age (30.3%), albeit the age range was wide and with good representation of the 3-40 yrs (21.6%) and 51-60 yrs age (20.4%). They corresponded to 22 nationalities being mainly from Spain, the UK, France and Germany.


They were previous visitors (92%) having sailed in Mallorcan waters more than 10 times (11-20 10.8%, 21-50 times 14.6%, >50 times 33%). The majority visited more than Mallorca Island (60.6%).

The boat was generally in the 10-15 m range (84%) mainly sailing boats (57%) and motor boats (36.8%).

Motivation

87% of the nautical tourists were motivated by relaxation, albeit to swim was equally important (86%) followed by enjoying nature (70.7%). Fishing was only relevant for 16.8% of the respondents. Accordingly, the selection of the place was motivated by shelter (4.5/5), good swimming (4/5), good views (4/5) and good fishing grounds be-ing the less important (1.5/5).

The relative impact of the RF of the nautical tourists could be relevant because 23% fished during their holiday. The main modality was trolled line (60%) and currican (48%). The mean expenditure was 278 € not considering the boat renting.

The impact of this activity could be relevant because if 23% of the 73997 touristic boats in 2009 fished the effort is important albeit not the main motivation of this sec-tor.

4.11.2 New methods tested in 2011

4.11.2.1 Classification of the recreational fishermen

The wide variability of the economic expenditure, as well as of the fishermen activity in general, causes a problem to extrapolate data with s.d. higher than the mean val-ues. Therefore the first step is to stratify the economic information to reduce the vari-ance. From the expenditure distribution obtained from the 285 enquiries with all economic data questions answered a negative exponential function with median 1051 € and confidence limits 38.4 and 5595.66 €, respectively, was evident.

Figure 24. Distribution of the expenditure (euros) by the number of answers.

The segmentation of the RF was performed using k-means in a pam function using R. The clustering was based on the euclidian distance. This technique implies the selec-tion a priori of the number of groups. The best number of groups was selected by means of a matricial correlation between the original Euclidian distances and the


calculated Euclidian distances when each object had the expenditure values from the representative object of its group. Therefore, when increasing the group number the explained variance increases till 1 (100%). The explained variance did not increase when considering more than 6 groups, then 6 groups were selected to classify the RF resulting in stabilized values.

Figure 25. Variance explained by groups numbers considered.

A PCA of the expenditures by group related to the activity and to the boat use, re-spectively, showed the groups being related to: RF with boats requiring small main-tenance expenditure, RF with boat in a port, RF with more expenditure in boat maintenance, RF which boat has required reparations, RF with high naval electronic expenditure in their boat, RF with high expenditure not related with the activity or the boat.


Figure 26. Grouping of the expenditures of recreational fishermen.

Economic impact of the recreational fishery

The impact was determined from input-output models using the data of the enquiries and considering the TIO tables published for Mallorca corresponding to 2004. The application of the different concepts in the MIOB tables was based on the TIO table albeit in some cases the correspondence of the kind of the expenditures was compli-cated.

A first result is the amount of money collected by the licensing that represents for 2009 336,022 € income for the administration. The economic incidence of the recrea-tive fishing into the Mallorcan economy (Cardona et al. in prep) showed a general expenditure of 49 million € in 2009, corresponding 12 to the activity and 37 million € to the expenditure linked to the boat use. This expenditure represents 0.185% of the GDP mainly implying the nautical sectors (>10%).

4.11.2.2 Measuring cpue using voluntary fishermen

The methods tested to determine the cpue proved to be unreliable due to unwilling-ness of the RF to let their bag to be weighed, bad recollection of data, and unrealistic reporting. Therefore we have started a series of collaborations with RF to determine their cpue by means of organizing outings with their participation. These tourna-ments are informal, based on the collaboration of volunteers that are rewarded by means of small presents, and include formation (talks by biologists and experts), fi-nalizing with some joint even like a barbecue. The aim is to obtain precise and accu-rate information and to build trust as baseline for future cooperation.


0.0

0.5

1.0

1.5

2.0

2.5

3.0

Maig P

rimer

Ai Bitx

o

Edgew

ater

Sansa

raMoro

Doncella

MajuSeri

ni

Lorem

ar

Portixo

lers

0123456789

CPUE angler/hourCPUE squid/angler

Figure 27. Loligo vulgaris capture per unit of effort on 10 different boats during an experimental fishing.

4.11.2.3 Evaluation of the catch composition and CPUE of the spearfishing modality within recreational fisheries of the Balearic Islands (W Mediterranean).

The number of spearfishers represents about 4% of the total recreational fishermen of the Balearic Islands (Grau, 2008). The in situ monitoring of this modality results diffi-cult from two main aspects: a) because of the atomization of the fishing effort (a sin-gle or a couple of spearfishers isolated in tens of coast-line kilometres) and b) because of the reticence of spearfishers to be interviewed.

In this context, we proceeded to focus our study in areas where this activity, both from the legal as from the catch point o view, is exhaustively under control by fishery authorities. In three out of seven fishery multiple-use MPAs managed by the local Government, spearfishing is allowed under particular conditions. This modality can be practised in the partially protected or buffer areas of these MPAs with a special license where all species and days allowed as well as bag quotas are detailed. The most overriding measures of this license are the ban of fishing for 3 days a week at each MPA or allowing the catch of only one piece for some very sought-after species.

The first question that arouse in front of this framework is whether fish stocks in these partially protected areas were in a comparable status respect open access areas, behind MPAs boundaries or, on the contrary, a greater catch should be expected be-cause of partial protection effect or even from spill-over from no-take areas at each MPA.

The results of the monitoring of fishery resources that the local Government is devel-oping in these MPAs since 2000 show that the state of fish populations is good on partially protected Posidonia oceanica seabeds and on patches of this seagrass with sand and rock. Nevertheless, fish populations have not recovered in the partially protected areas where rocky botoms are dominant, just the habitat type where spear-fishing is effective.

In the light of these results, using spearfishing CPUEs from rocky partially protected areas to estimate the overall impact of this modality in the Balearic Islands could be a realistic approach. During 2009 and 2010 the surveillance services of Palma, Migjorn and Llevant MPAs initiated a data record process on spearfishing activity. Each spearfisher sought was requested for the license, his catch was measured and weighted and interviewed with 5 main questions: a) the time spent fishing (in the cases that his/her arrival was not detected), b) the number and wet weight of fish caught for each different species (directly measured by the rangers), c) had he/she


arrived from the shore or by boat to the fishing point, d) the depth he/she were fish-ing on, e) whether he/she had participated in any championship in the last five years.

By such a protocol we undertook 347 interviews. Among these, 296 showed data about catch and time of fishing and could be used to make a preliminary description of this activity. Two hundred and seventeen records out of 296 showed some catch and 79 had zero catch. Thirty three species were recorded with a total amount of 653 fish weighting 484 kg. Only three species which are also targeted by the commercial fleet accounted for near 50% of the total catch: Octopus vulgaris, Diplodus sargus and Epinephelus marginatus.

Significant differences were found among different MPAs; Llevant and Migjorn showing greater CPUEs than Palma MPA. The total gross averaged CPUE was 0.568 kg x fishermen-1 x hour-1. As a first approach and taking into account published data about mean time of fishing, fishing daily frequency (Morales-Nin et al., 2005) and the estimated number of spearfishers from registered licenses in the local administration, this modality could account for up to 10% of the 1,200 tones that recreational fishing are actually harvesting in the Balearic Islands. These must be considered as prelimi-nary results given that new data were incorporated at the end of 2010 and some is-sues as the minimum time considered to calculate CPUE should still be refined.

4.12 Sweden

For more information concerning the general overview of the national recreational fisheries and on the detailed description on national recreational fisheries, see the 2010 PGRFS report.

4.12.1 Possible sampling frames

National mail screening survey in 2011

There has been no national mail screening survey in Sweden since 2009 (regarding recreational fisheries in 2008). A new national mail screening survey is carried out during spring 2011 regarding recreational fisheries 2010 and will be reported during 2011.

The 2011 survey is different from the 2009 survey:

Number of questionnaires is the same as in the 2009 survey (10 000) however in 2011:

1 ) We re-used 2000 addresses from all responding person who answered that they were recreational fishermen in the 2009 national mail screening sur-vey

2 ) 8000 new addresses were selected randomly from the government address list of persons living permanently in Sweden.

3 ) Of the 8000 new addresses about 1500 persons were selected geographi-cally from the very southern part of Sweden. The reason for this was to achieve a better coverage of recreational fishery on cod in the part of the Baltic Sea were cod especially occurs.

4 ) The questionnaire has been changed marginally;

a) Questions on number of fishing days abroad and in which country have been included.

b) Questions on recreational fisheries from charter vessels in the Sound have been included.


Surveys on recreational fisheries targeting cod.

The Sound is the most important recreational catch area in the Baltic Sea regarding catches of cod. We are therefore interested in obtaining better quantitative data from data area. New sources of information are:

a) A captain on a charter vessel in the Sound has in a daily questionnaire (6 month in 2010) reported the number of recreational fishermen onboard and the catches of cod (kg). The daily questionnaire has been compared to the vessels daily blog on the website with information on number of fishermen and catches of cod. Information from the rest of the skipper’s blogs (totally ten vessels) on the vessels websites has been compared with the reported catches from the captain’s questionnaire. The total catch of cod taken from charter vessels in the Sound has been calculated from these sources and other information about the recreational fisheries from the charter vessels. The re-sults indicate that the catches in 2010 was 40 – 80 tons.

b) During 2011 a crew member on each charter vessel (9-10 in total) operating in the Sound will visually estimate the total catch of cod in kg during each tour. Only cod that is kept are included. Personnel from the Institute of Marine Re-search will visit each charter vessel on one occasion during 2011 and take the actual weight of cod catches in order to compare those weights with those es-timated by crew. Table 28 shows the results for the first 3 months of 2011. If the total catch in table 28 is recalculated for 12 months we obtain approxi-mately 60 ton per yr.

c) Data on recreational cod fisheries from the national mail screening survey 2009 is so far used providing data on catches of cod in 2010. The national mail screening survey 2011 contains new questions about recreational fisheries from charter vessels in the Sound. The results from the national mail screen-ing survey will be used as a cross-check and compared with the above stud-ies on the charter vessels in the Sound. If the results are similar we may accept the national mail screening survey data on cod catches from small boats and from the shore in the Sound.

Surveys on recreational fisheries targeting salmon.

The most common recreational fishery for salmon is:

• Angling in salmon rivers target salmon and sea trout and in some rivers also grayling.

• Offshore trolling in south Baltic target salmon and sea trout. • Net, seine fishery and fishery with trap-nets by fishermen were until 2010

recorded as recreational fishery.

The compilation of river statistics is difficult due to the varying degree of organiza-tion and different kinds of fisheries occurring in different rivers. In addition the vary-ing size of the rivers gives rise to a need for variable approaches. In almost all rivers angling is covered by data from individual fishery rights owners, but the quality of this statistics is very variable. In most cases there is a requirement of fishermen to report their catch, but this rule is enforced to a variable degree. Data on other kind of fisheries (seine, net, trap-nets in rivers) are normally collected via questionnaires from authorities to individual fishermen. Offshore trolling has been studied by screening and list frame surveys, complemented by data from statistics from major


ports. Even though the quality of data in some cases is of high quality, no statistical methods have been used to estimate the precision and accuracy of the data.

The results from the screening survey conducted by mail questionnaires will be used as a cross-check and compared with other surveys on recreational salmon catches.

4.12.2 Available statistics

Catches of cod in the Sound by licensed professional fisheries

Since 1932 trawl fisheries is forbidden in the Sound. The major part of the catches of cod is therefore taken in net fisheries. The reported (minimum) catch of cod by pro-fessional fishermen was 2008-2010 in average 406 tons per yr.

Cooperation with NGO and volunteers

In on-site surveys and in some cases list frame surveys have NGO and volunteers been involved. Public funding has partly been used by volunteers and NGO.

Table 28. Total Swedish cod catch in kg during tour boat fishing between Jan-March 2011 in the Sound (between Sweden and Denmark

Ship Kg cod1 X 41512 X 35053 x 11954 X 13895 X 15596 x 8357 X 13928 not operating9 X 93110 X 1925

TOT SUM Kg 16882

19188

Tot n toursMean kg cod /tour

SUMMARY: Preliminary results until March 31. Based on visual estimations made by crew during each cruise. Data from 9 ships are included. 1 ship was not in operation for more than a few days. A fishing tour normally last for 4-7 h. Cod catch taken by foreigners are included = 3% if total catch (554 kg). Some of these foreigners may be residents of Sweden. Control of cod weight during 10 fishing tours will be done during 2011.


Table 29. Reported landing of cod in the Sound (subdivision 23) by licensed professional fisher-men, gillnet fishery.

Year Landings (tons)2000 6902001 6802002 4502003 5302004 3902005 6502006 6402007 4902008 5002009 2602010 440

Table 30. Salmon recreational catches in the Baltic Sea

Year Weight (kg) Number Weight

(kg) Number Weight (kg) Number

2008 69568 13809 106941 17860 176509 316692009 88205 18248 73650 12697 161855 309452010 70045 12827 33351 5183 103396 18010

Trapnet, angling, Seine net, Gillnet

Unlicensed (Angling, Trapnet, Seine net, Gill

net )

Total Baltic Sea

Baltic Sea Rivers emptying in Baltic Sea

Recreational catches of salmon in the Baltic Sea is partly known.

Recreational catches in rivers emptying the Baltic Sea is in some rivers well known and in other rivers only partly known


Table 31. Salmon recreational catches in the Kattegat and the Skagerrak

Year Weight (kg) Number Weight (kg) Number Weight

(kg) Number

2008 approx 500 140 15000 3750 15500 3890

2009 approx 500 140 14000 3500 14500 36402010 approx 500 140 20000 5000 20500 5140

Kattegat/Skagerrak Rivers emptying in Kat/Skag.

Unlicensed (Angling, Gill net )Angling, Gillnet

Total Kat/Skag.

Recreational catches of salmon in the Kattegat and the Skagerrak are nearly un-known. Recreational catches in rivers emptying in the Kat/Skag are very well known.

For other cod and salmon data available in Sweden, see the PGRFS 2010 report

4.13 U.S.A

4.13.1 Updates on the U.S. Marine Recreational Information Program

The Marine Recreational Information Program (MRIP) is continuing to make progress toward developing and implementing improved surveys designs for monitoring marine recreational fishing effort and catch. This report briefly describes some of the findings of recently completed or current projects that should be considered in devel-oping or improving survey designs to serve a similar purpose in European countries.

4.13.2 Matching Estimation Methods with Sampling Designs in Access Point Surveys

One of the highest priorities for the MRIP has been to develop improved estimation methods that properly account for the complex sampling designs used in current and past surveys of marine recreational fisheries. In 2004, NOAA Fisheries contracted with the U.S. National Research Council (NRC) to conduct a critical review of U.S. recreational fishery survey methods, and the NRC published a report of their criti-cisms and recommendations in 2006. One of the major criticisms was that most of the on-site surveys of fishing access points were using simplified estimators of total fish-ing effort and/or mean catch rates that ignored complexities in the sampling and were, therefore, likely to be biased. One prominent example is the Access Point An-gler Intercept Survey (APAIS) that has been conducted as a component of the Marine Recreational Fishery Statistics Survey (MRFSS) since 1979. The sampling for that sur-vey has been conducted using what can be described as a stratified, multi-stage clus-ter sampling design. However, the estimation method used has assumed a simple random sampling design. In addition, the sampling for the APAIS was designed to allow considerable flexibility in the temporal and spatial distribution of sampling effort in order to maximize the amount of data collected per hour of sampler effort. This added even more complexity to the sampling protocols and increased the risk of sampler errors. The NRC stated that improvements were needed in both the estima-tion methods and the sampling protocols to reduce the potential for bias in point estimators of mean catch per angler trip and provide more accurate estimators of the precision of those point estimators.


The MRFSS is a complemented telephone-access design that utilizes a Coastal Household Telephone Survey (CHTS) to estimate the number of angler fishing trips and the APAIS to estimate the mean number of fish caught per angler fishing trip. The CHTS only covers residents of counties on or near the coast (coastal residents), so the APAIS is also used to estimate the proportion of fishing trips made by costal resi-dents with telephones and use that proportion to estimate the numbers of fishing trips made by non-coastal residents and coastal residents without telephones. The fishing effort estimates from the two surveys are combined to produce estimates of total fishing effort that are then used to expand APAIS estimates of mean catch per angler trip.

In order to ensure that both the estimators of total fishing effort and mean catch rates are unbiased, the NRC recommended the use of properly “weighted estimators” that accurately account for the complex sampling design. In order to do that, it is neces-sary to determine the sample inclusion probability for each intercepted angler fishing trip, use the inclusion probability to calculate an appropriate sample weight for that trip, and apply that weight to the trip data in the estimation process.

The NRC also recommended eliminating much of the flexibility that was built into the APAIS sampling protocols. Implementation of more rigid protocols would make it much easier to determine the correct inclusion probabilities for all sampled angler fishing trips and would also reduce the risk of sampler errors.

4.13.2.1 Needs for Implementing Weighted Estimation

Weighted estimators can be derived very easily as long as the APAIS sampling de-sign is known and the data needed for the calculation of trip inclusion probabilities are collected. The sampling for the APAIS follows a multi-stage cluster sampling design with a completed angler day of fishing as the ultimate sampling unit. The sampling is stratified by fishing mode (shore, private/rental boat, or charter boat), geographic area (state in most cases), month, and day type (weekend/holiday or weekday). The primary sampling unit (PSU) is a site-day combination that is selected from a spatiotemporal frame that is basically a matrix of available fishing access sites and available fishing days within a given stratum. The secondary sampling unit (SSU) differs by fishing mode. In the boat modes, the SSU is the cluster of boats re-turning from a day of fishing to the selected PSU. In the shore fishing mode, the SSU is the cluster of anglers who completed a day of fishing at the selected PSU. The terti-ary sampling unit (TSU) in the boat modes is the cluster of anglers who have com-pleted a day of fishing on the selected boat, or SSU. The cluster of fish landed by an intercepted shore angler is sampled for size measurements (length and weight) in the tertiary stage, while the cluster of fish landed by a boat angler is sampled for size measurements in the quarternary stage.

The inclusion probabilities for each intercepted angler fishing day, or sampled fish, must be based on the calculated selection probabilities at each stage of sampling in the sequence that leads to that intercept. For example, suppose we are selecting sam-ple within the weekend/holiday stratum of July for the private/rental boat mode in North Carolina. Let’s say there are 30 available access sites and 8 available days for sampling. If each site/day combination has the same selection probability and we draw a total sample of 20 site-days, then the inclusion probability for any one site/day is 20*(1/240), or 0.08333. For one selected site-day, there may be a cluster of 25 boats that return from fishing over the course of the day, but the assigned sampler only intercepts 5 of those boat trips. The inclusion probability for one of those trips in the secondary stage of sampling would be 5/25, or 0.20. For one selected boat trip, there


may be a cluster of 4 anglers who fished on the boat that day and only 2 of them are successfully interviewed. The inclusion probability for each of those two anglers at the tertiary sampling stage would be 2/4, or 0.50. The product of the inclusion prob-abilities calculated at each stage gives you the total inclusion probability for one of those selected angler fishing days as 0.08333 X 0.20 X 0.50 = 0.008333. The inverse of that probability is the appropriate weight to apply to data collected for that angler day-trip in the estimation process.

A further complication is that the selection of site-day assignments for the APAIS is made with “unequal probability sampling”. The selection probability of a given site is based on the expected fishing activity level, or “fishing pressure”, at the site. Sites with greater activity are given a higher probability of being selected for sampling than sites with lower activity. This type of sampling can also be described as “prob-ability-proportional-to-size” sampling, where the size measure is the measure of fish-ing pressure.

In order to weight the site-specific data obtained with PPS sampling, it is important to take the actual site selection probabilities into account. In other words, the angler trips intercepted at higher pressure sites need to be “weighted down” to account for the fact that they had a greater probability of being included in the sample than trips intercepted at the less frequently selected lower pressure sites. The appropriate sam-ple weights can be easily calculated as long as the selection probabilities are known for all sites in the sampling frame.

Although the APAIS has always selected site-day assignments through a randomized process such that the selection probabilities for those PSUs can be easily calculated, the sampling protocols have also allowed samplers to visit up to two additional “al-ternate” sites during an assignment if the number of completed angler fishing trips encountered is likely to be very low at the assigned site. Samplers leaving to visit alternate sites have been instructed to always visit the next nearest site expected to have fishing activity in the assigned fishing mode, but it is not clear that all samplers have consistently implemented this approach. Different samplers have had different interpretations of which sites are the “next nearest”, and some samplers assumed that they should only go to alternate sites that are certain to be productive on the given day assigned. Needless to say, it is not easy to determine the inclusion probability of any given site as an “alternate site”. Without some measure of this probability it is difficult to determine how to appropriately weight the angler trip data collected at the alternate sites.

As explained in the simple example described a few paragraphs earlier, the calcula-tion of appropriate sampling weights for the secondary, tertiary, or quarternary stages of a multi-stage sampling design can be straightforward as long as accurate cluster counts are obtained at each stage. Any selected boats, anglers, or fish that represent only a sample of a known cluster, should be “weighted up” to represent the whole cluster. The weight is simply calculated as the inverse of the sampling fraction.

The APAIS has always collected counts of angler trips ending at the assigned and alternate sites that were not intercepted. By adding the count of missed angler trips to the number of intercepted angler trips, it is possible to calculate the total number of angler trips encountered by the sampler. However, in the private/rental boat and charter boat strata the APAIS has not collected counts of missed boat trips. Therefore, total angler trip cluster sizes are measured, but the distribution of those clusters across different boat trips are not. This makes it difficult to determine the appropriate


secondary and tertiary stage sampling weights for intercepted angler trips in the boat modes.

A further complication is that the APAIS has instructed samplers to pick the most productive time of day to visit an assigned site and has not required the samplers to stay on site for a standard length of time. As a result, the time of day and duration of sampling at each site can be quite variable from one sampling assignment to the next. Since the time unit for the site-day sampling is actually a whole day, it is necessary to expand the cluster counts obtained on site to represent the total cluster size for the corresponding 24-hour period. This can be very difficult given that all of the APAIS sampling has been focused on collecting data during only a brief period when activ-ity was likely to be near its peak. Little or no count data has been obtained during off-peak hours.

4.13.2.2 Modeling to Support a Weighted Estimation Approach

The MRIP Sampling and Estimation Work Group (SEWG) formed a project team supported by two expert consultants to develop suitable modeling approaches that could be used to develop an appropriate weighted estimation method that would be applicable to the APAIS data. Drs. Jay Breidt and Jean Ospomer at Colorado State University were hired because of their known expertise in survey design and estima-tion.

As explained above, the probabilities for sites to be selected as “alternate sites” in the APAIS are not known directly from a formal randomized sample draw process. Rec-ognizing that a large amount of data has been obtained by allowing samplers to visit alternate sites, the project team reasoned that it would be desirable to find a way to appropriately weight these data to support design-unbiased estimators for the APAIS. Therefore, a modeling approach was developed to approximate “pseudo-weights” for alternate site-day samples. Because sampler visits to alternate sites are contingent on what happens at an assigned site that was selected through a formal sample draw, the team first considered trying to model the sampler decision-making process to develop contingent probabilities for sites to be selected as alternate sites based on proximity to drawn sites and observed activity at the drawn sites. This proved to be too difficult mainly because it was not clear that all samplers had consis-tently adhered to the rules for selecting alternate sites or had interpreted those rules in the same manner. For this reason, the team decided to use the historical frequen-cies of sites being selected as alternate sites over several years to model their inclu-sion probabilities in any given year of that time series. The inverse of the modeled inclusion probability for any given site to be selected as an alternate was then used to assign an appropriate “pseudo-weight”. The weight for any given site could then be calculated as a combination of its weight based on the formal sample draw and its pseudo-weight based on the informal selection of alternate sites by the samplers.

Another challenge was to determine a way to estimate the number of completed boat trips that samplers could potentially have intercepted at each sampled site. Samplers have always been required to record counts of “missed” angler fishing trips ending at sampled sites, but they were not required to record counts of “missed” boat trips. This makes it difficult to determine the appropriate sampling fractions to use for cal-culating sample weights at the second stage of sampling in the private/rental boat and charter boat fishing modes. Fortunately, since the early 1990’s the APAIS has required samplers to record the number of anglers who fished on the same boat trip for every angler that is interviewed. The team decided to use these data to estimate


the mean number of anglers per boat trip for each sampled site-day. The inverse of this estimated mean could then be multiplied by the total count of missed angler trips to get an estimate of the number of missed boat trips. That estimate was then used along with the number of intercepted boat trips to represent the number of boat trips encountered during the time the sampler was at the site.

In order to get valid measures of the boat trip clusters and angler trip clusters for each sampled site-day, the team had to determine a way to expand the measures obtained for the time interval when the sampler was at the site to get a result that would represent a full 24-hour day. The APAIS has not required samplers to stay at an assigned site for a full day, and has also not required them to stay for any stan-dard length of time. This makes it very difficult to know how to expand the obtained counts of angler trips and the estimated counts of boat trips during any given sam-pled time interval. The team looked for other sources of data that could be used to approximate an expected distribution of fishing trips across a 24-hour interval at any given site. The best source available is the CHTS. In that telephone survey, respon-dents have always been asked to report the time of return for each fishing trip they report. The team reasoned that a time series of CHTS data on trip return times for each fishing mode and month could be used to model the expected temporal distribu-tion of trips for full day in any given month. The boat trip and angler trip counts obtained in any particular “time slice” covered by a sampler could then be expanded using the modeled temporal distribution of trips to estimate the cluster sizes for a full day. These expanded full-day cluster sizes could then be used to determine the ap-propriate secondary and/or tertiary stage weights to apply to the data collected for any given site-day sample.

It was very difficult to determine an appropriate way to weight alternate mode angler trip intercepts that were obtained opportunistically. Modeling approaches were con-sidered but determined to be too complex. Since alternate mode intercepts only com-prised 5-15% of the total number obtained by the APAIS for any given fishing mode, the team decided not to use any alternate mode intercepts in the weighted estimation.

In summary, this new weighted estimation approach for the APAIS is design-unbiased. It uses site-day sample selection probabilities to weight the data and prop-erly takes the multi-stage cluster sampling design and associated stage-by-stage in-clusion probabilities into account. Assigned site-day sample probabilities are known, and alternate site-day sample probabilities are approximated using a modeling ap-proach. Available data on cluster sizes at each stage of sampling are used to deter-mine cluster totals for sampled time slices at each sampled site, and a modeling approach is used to expand those time slice totals to determine the cluster totals for the full 24-hour day. The estimated 24-hour cluster totals are used to calculate appro-priate sample weights for secondary and/or tertiary stages of sampling. Opportunis-tic samples of angler trips in fishing modes other than the assigned fishing mode are not used in the APAIS estimates of mean angler catch rates.

The new weighted estimation method combines design-based methods with model-based methods to achieve the desired objective. The uses of sample selection prob-abilities and cluster sampling fractions to weight sample data are design-based meth-ods that are textbook best statistical practices. The adjustments made for estimating alternate site selection probabilities and expanding time slice cluster counts to full day totals are model-based approaches that required novel statistical procedures.

The team prepared a report on this new estimation method, and the report has sub-sequently been subjected to three external peer reviews. The report and the team’s


written response to the peer reviews is available on the MRIP website (www.countmyfish.noaa.gov/).

The MRIP is now using the new weighted estimation method for APAIS data to re-estimate marine recreational fishery catches for 2004-2011. The new MRIP catch esti-mates will be compared to the original MRFSS estimates for those years to assess the direction and magnitude of any potential biases in the original MRFSS estimation method. Final results are currently under review and will be released to the public in early 2012.

Preliminary results show that for most finfish species, the MRIP estimates are not consistently higher or lower than the MRFSS estimates over the time series, and the differences for any one year are not statistically significant. There are consistent dif-ferences for some species, but for the most part those differences are not large enough to be statistically significant for any one year. However, for some of those species the mean differences over the time series are statistically significant. For a few species, the MRIP estimates are consistently higher than the MRFSS estimates, and for a few the MRIP estimates are consistently lower. The differences between MRIP and MRFSS estimates are largely attributable to the differential weighting of trips sam-pled at sites with different fishing activity levels. The MRFSS sampling design gives angler trips ending at the more active sites a much higher probability of being in-cluded in the sample than angler trips ending at the less active ones. The weighted estimation method gives the trips intercepted at the more active sites a lot less weight than the trips intercepted at the less active ones, but the original MRFSS gave all in-tercepted trips the same weight in the estimation process. For a few species, it ap-pears that there are relatively consistent differences between high-activity and low-activity sites in the mean catch rates of intercepted angler fishing trips. For those spe-cies, the original MRFSS estimator of the mean catch per angler trip appears to be significantly biased.

Preliminary results clearly indicate that the original MRFSS estimator of the variance of any given APAIS point estimator is much too low. The new weighted variance estimators provide much more realistic measures of the precision of the resulting point estimates of catch rates and total catch. In short, the original MRFSS variance estimators were grossly incorrect.

A very important conclusion drawn from this study is that the relative precision of an access point survey estimator depends more on the number of site-days sampled than on the total number of angler trips intercepted. The important “sample size” target to increase for improving precision is the number of primary sampling units (site-days), rather than the number of ultimate sampling units (completed angler fishing trips). Future access point surveys should be focused on maximizing the number of site-days sampled with available resources.

4.13.3 New Access Point Survey Sampling Design

The MRIP has also been working with expert consultants to develop an improved sampling design for future access point intercept surveys that would fully support a design-based estimation method and significantly reduce the potential for non-sampling errors. A project team consisting of four expert consultants and representa-tives of NOAA Fisheries and three different state agencies was formed in 2009 to develop a new sampling design. The consultants include Drs. Jay Breidt and Jean Opsomer from Colorado State University and Dr. James Chromy and Breda Munoz from RTI, Inc. A pilot study was conducted in North Carolina in 2010 to test the new

http://www.countmyfish.noaa.gov/


design in a side-by-side comparison with the original MRFSS APAIS sampling de-sign. The results of the pilot study are currently being analyzed and further im-provements in the sampling design to improve its efficiency and cost-effectiveness are being considered. NOAA Fisheries has begun preparations to implement im-proved access point surveys on a broader scale in 2013, and the recommendations of this project team will help to finalize the plans for full implementation.

The new sampling design differs from the traditional MRFSS APAIS design in several ways that will ensure more accurate accounting of the site selection probabilities and stage-by-stage cluster sampling fractions needed to support a fully design-based weighted estimation approach. The new design does not allow samplers the discre-tion of selecting alternate sites for intercepting and interviewing anglers. Site clusters are formed prior to a formal sample draw process, so that samplers can visit more than one site during a given day and all visited sites will have a known selection probability. Sites are clustered based on their proximity to one another and their es-timated levels of fishing activity. If a site has an estimated fishing activity level that exceeds a certain threshold, it is not clustered with another site and is eligible to be drawn as a “1-site cluster”. If a site below that threshold is paired with the next near-est site and the total activity level estimated for the two sites exceeds the threshold, then those two sites are kept together as a “2-site cluster”. Otherwise, a maximum of 3 proximate sites are clustered together into a “3-site cluster”. The formation of site clusters is done using an ArcGIS algorithm that minimizes driving time between sites and applies the threshold rule to determine the number of sites in a cluster. The total fishing pressure of each formed site-cluster unit is used as its weight in the PPS sam-pling of site-cluster units, and the order in which sites within the cluster are to be visited is randomized and fixed prior to issuing the interviewing assignment. This approach allows for easy calculation of the inclusion probabilities for each site within a selected site cluster.

To simplify the determination of appropriate cluster sampling fractions, sampling is stratified among four 6-hour time blocks and samplers are assigned a consistent time interval for sampling at each assigned site. Drawing assignments for six-hour time blocks that are outside of the period of peak fishing activity ensures that at least some sampling is occurring at all different times within a full 24-hour day. This will also remove any potential bias that may result because nighttime or off-peak daytime fishing trips tend to have higher or lower catch rates than the peak daytime fishing trips that have traditionally been the focus of MRFSS APAIS sampling. Sampling across the four time blocks does not have to be equal and can be optimally allocated to improve the efficiency of sampling and the precision of resulting survey estimates.

The consistent 6-hour time interval for sampling eliminates the need for modeling some sort of expansion of cluster counts for a time slice of indefinite duration. Sam-plers are given specific instructions to stay at each assigned site for a given length of time. When assigned to a three-site cluster, the sampler is told to spend 2 hours at the first site, drive to the second site, spend 2 hours at the second site, drive to the third site, and spend the remainder of the 6 hours there. When assigned to a two-site clus-ter, the sampler is told to spend 3 hours at the first site, drive to the second site, and spend the remainder of the 6 hours there. When assigned to a single site, the sampler is instructed to spend the whole 6 hours at that site. In this design, the PSU can be easily subdivided in the estimation process into well-defined set of spatiotemporal units for which appropriate weights can be easily calculated. Also, there is no need for expanding any counts to estimate totals for unobserved time periods.


The new sampling design places a much higher emphasis on maximizing the sam-pling of site-days and getting accurate counts of missed boat and/or angler fishing trips. A formal sample draw program was developed that takes staffing constraints into account and ensures that all randomly selected sets of assignments can be cov-ered by the existing staff. This is an important improvement because it will greatly reduce the amount of unobserved sample, or sample nonresponse. The MRFSS APAIS has traditionally drawn more assignments than can actually be completed, allowing for the possibility that there could be some bias in the way drawn assign-ments were selected for completion. The only assignments that would not get com-pleted under this new design would be those not done due to unexpected circumstances such as sampler illness. In addition, samplers are instructed to balance their work on site between conducting angler interviews and obtaining accurate counts of boats and/or anglers at the site that have just finished fishing for the day. This will ensure that cluster sampling fractions can be accurately calculated and used in the weighting of intercepts to accurately account for their inclusion at each stage of sampling.

Although the pilot study version of the new sampling design targeted angler fishing trips as the ultimate sampling unit in the boat mode sampling strata, the project team has proposed collecting catch data at the boat trip level in future access point surveys of private/rental boat and charter boat fishing. This is a change from the prior MRFSS angler-based approach that would eliminate a stage of sampling and improve the precision of mean catch rate and total catch estimators. It will probably still be neces-sary to get data on the number of anglers who fished on sampled boat trips so that mean catch rates can also be calculated at the angler level and matched with esti-mates of the total number of angler trips from off-site surveys.

A very important feature of this new sampling design is that it allows for more de-sign-based estimation and eliminates the need for the modeling components used in weighting MRFSS APAIS data.

4.13.4 New Dual-Frame Sampling Designs for Surveys of Fishing Effort

The 2006 NRC Report pointed out a number of issues with the Coastal Household Telephone Survey that has been used by the MRFSS to estimate recreational fishing effort on private/rental boats, man-made shore structures, and natural shorelines. Off-site sampling methods that rely on telephone interviews are complicated by the increasing use of cellular telephones, answering machines, and caller ID. In general, telephone survey response rates are declining at a rapid rate and increasing propor-tions of households do not have, or do not use, landline telephones for communica-tion. In addition, the NRC Report specifically stated that the “existing MRFSS random digit dialing survey suffers in efficiency from the low proportion of fishing households among the general population and may allow bias in estimation from its restriction to coastal counties.” The NRC recommended the development and use of fishing license-based lists as sample frames for more efficient off-site surveys, but also recognized that dual-frame procedures will be needed to avoid potential undercover-age biases as long as available lists of fishing participants are incomplete.

MRIP has been conducting a number of studies to investigate better designs for off-site surveys of fishing effort. Pilot studies conducted in North Carolina and Louisiana have shown that license-based telephone surveys of known fishing participants can be much more efficient than RDD surveys. However, the recreational fishing license lists used in both states are incomplete. Not all active anglers are currently required to obtain a license, and many license holders do not provide valid phone numbers.


The results indicate that any license-based approach would have to use a second, more complete, household sampling frame to reach the participants who can’t be reached by sampling from the license-list frame. A dual-frame approach would pro-vide improved sampling efficiency while assuring relatively good coverage, but such an approach is complicated by the need to accurately determine the overlap between the two frames. It is necessary to know the overlap, so that survey estimators can be developed that will prevent double-counting of fishing trips reported by respondents who could be reached through both sampling frames. The overlap can be determined by matching telephone numbers selected in the RDD sampling with those listed in the license frame, but many households with licensed anglers have more than one landline telephone number and licensed anglers with landline phones may provide cell phone numbers when purchasing a. Therefore, it can be very difficult to deter-mine the overlap domain and incorporate it properly in the estimation process.

Because dual-frame telephone surveys can be problematic and would still be subject to declining coverage and response rates, an MRIP project team was formed to exam-ine the feasibility of using a dual-frame mail survey approach. The expert consultants recruited to participate on that project team include Dr. Lynne Stokes (Southern Methodist University, also an NRC panel member), Dr. Mike Brick (Westat, Inc.), and Dr. Nancy Mathiowetz (University of Wisconsin - Milwaukee). The team developed a design that utilizes a fishing license, or fishing registry, list in conjunction with a rela-tively complete list of residential mailing addresses that is now available for purchase by survey research firms. The latter list is the “Delivery Sequence File” (DSF) main-tained by the U.S. Post Office. Through address-based sampling (ABS), the DSF pro-vides access to more residential households than could be reached through an RDD telephone survey, including those with no landline telephones. The survey of the DSF is done in two phases, but the survey of the license frame is done in one phase. In the former, a sample of mailing addresses is randomly selected for a very brief screener survey. Mail questionnaires are sent to the selected addresses with a minimum of questions needed to identify if any household residents made any marine recrea-tional fishing trips in the past year. Those households that respond positively are then included in a second phase sample who are mailed a more detailed question-naire asking about the numbers of shore and private/rental boat fishing trips made in the two-month reference period. An initial pilot study was conducted in North Caro-lina in 2010, and that study demonstrated that a dual-frame mail survey approach could obtain higher response rates (~40%) than the current RDD telephone survey of the MRFSS (~25%) in North Carolina. In addition, the use of ABS allows for much more accurate frame-matching to identify the domain of overlap in the dual-frame sampling. It is much easier and more reliable to match license holder mailing ad-dresses to those selected from the DSF than it would be to match their listed phone numbers with those generated in the RDD sample.

The pilot study described above also found evidence of what appears to be a non-response bias in the address-based sampling of households using the DSF. It was possible to match mailing addresses of households selected from the DSF with ad-dresses provided by license holders to identify sampled households with license holders. The mail survey response rates for those households were then compared with those for households not known to contain license holders. The known license-holder households had significantly higher response rates than the other households in the sample, indicating that households with fishing participants are more likely to respond. Because it is possible to match addresses between frames, it is actually pos-


sible to estimate the error associated with this apparent nonresponse bias and correct for it in the estimation process.

A new dual-frame study is planned to start in 2012, and this study will provide a side-by-side comparison of mail and telephone contact modes for both license frame and household frame sampling. The study will be conducted bimonthly in the South Atlantic subregion (North Carolina, South Carolina, Georgia, and the east coast of Florida) for a full year. The household samples will be selected from the DSF using an ABS approach. Half of the household sample will be used for a two-phase mail sur-vey, and the other half will be used for a two-phase telephone survey. In the latter case, a phone number will be determined for each selected mailing address by doing reverse look ups using available phone directories. Half of the license-holder sample will be used for a mail survey, and the other half will be used for a telephone survey. This design will allow for the use of mailing addresses to determine the overlap do-mains between alternative frames. In addition, it will allow for evaluation of four possible combinations of frame and contact mode in a dual-frame survey approach. The possible combinations are as follows:

1 ) Mail survey of license frame / mail survey of household frame 2 ) Mail survey of license frame / telephone survey of household frame 3 ) Telephone survey of license frame / mail survey of household frame 4 ) Telephone survey of license frame / telephone survey of household frame

Comparisons will look at response rates, possible mode effects (mail vs. telephone), survey costs, coverage of alternative household frames (RDD vs. ABS), and levels of nonresponse. The MRIP hopes to establish the best of these designs to implement more broadly in 2013 as the new survey design for monitoring shore and pri-vate/rental boat fishing effort on the Atlantic coast and in the Gulf of Mexico.

4.13.5 Other Ongoing MRIP Studies

There are a number of other MRIP pilot studies ongoing, and progress reports on those studies are being posted periodically on the MRIP website. There are a couple of studies that are particularly pertinent to the current work of the PGRFS.

4.13.5.1 Telephone-Diary and Web-Diary Surveys of Both Catch and Effort

A new pilot study was developed and started in 2011 to test the use of both tele-phone-diary and web-diary panel survey approaches for collecting data on marine recreational fishing effort and catch from known fishing participants. Two expert consultants, Dr. Virginia Lesser (Oregon State University) and Dr. Nancy Mathiowetz have helped in the development of the study and will be assisting in the evaluation of results. The design is based largely on the telephone-diary survey approach that has been used very successfully in Australia. The study is being conducted in both North Carolina and Florida to see if this approach would provide a good way to evaluate differences between the fishing trips reported at different times of day at public and private access sites. The access point surveys conducted in the U.S. are unable to col-lect data on angler fishing trips that end at private docks and shorelines, or at private locked marinas or boat ramps. In addition, they have not traditionally collected data on fishing trips that end during nighttime or off-peak daytime hours. Therefore, an off-site survey method is needed to get data that can be used to evaluate any poten-tial biases that may be caused by the omission of private access, nighttime, or off-peak daytime fishing trips. The panels for these surveys have been recruited through a dual-frame mail survey like the one described above for the North Carolina dual-


frame pilot study. Fish guides, angler diary forms, and detailed instructions for using both were mailed to the panelists. Panel members were asked to record data on their fishing trips and catch in the diary form so they would be able to easily recall this information when asked to provide it. Half of the members of each panel are being contacted through periodic telephone contacts to collect their fishing effort and catch data. These panelists comprise the “telephone-diary” test group. The other half were asked to report their effort and catch data through a website reporting process, and are identified as the “web-diary” test group. If the web-dairy members fail to report through the web, they receive a telephone call reminder. For the telephone-diary group, the telephone contacts are every two weeks for the more avid anglers and once a month for the less avid anglers.

4.13.5.2 For-Hire Boat Survey Methods

In 2008, the MRIP For-Hire Survey Work Group recruited three expert consultants – Dr. James Chromy, Dr. Ray Webster (International Pacific Halibut Commission), and Dr. Stephen Holland (Florida State University) – to conduct a thorough review of the current survey methods used in all U.S. states to monitor fishing effort and catch on charter boats and headboats (sometimes called “partyboats” or “open boats”). The report produced from that review is posted on the MRIP website. That review identi-fied a number of “best practices” for surveying the for-hire recreational fisheries. The consultants emphasized the need for developing complete registries of for-hire boats that can be used to develop and maintain a complete, current sampling frame of boats with needed contact information included for the owners and operators of those boats. Periodic contacts of boat representatives should be used to accurately identify landing site data for each boat.

The consultants also recommended the implementation of mandatory logbook census reporting programs for for-hire boats wherever feasible. If possible, electronic tech-nologies should be implemented to allow boat operators to file their reports elec-tronically. Ideally, boat owners should be required to complete a logbook report for each fishing trip that records specific information on fishing effort, location and methods, as well as specific data on landed and discarded catch of different species or species groups. The trip reports should be submitted weekly, and a significant com-pliance monitoring effort will be needed to make sure that reports are consistently submitted on time. It will also be important to allocate a significant effort to enforce-ment of the mandatory reporting requirement to ensure that all boat owners take the requirement seriously.

The consultants emphasized the need to validate the self-reported data provided in logbook reports by conducting an independent sampling survey that could be used for purposes of comparison and possible evaluation of any consistent response errors. For this purpose, they recommended the maintenance of a landing site frame that could be used for an access point survey of boat fishing trips. The fishing pressure for each landing site (in terms of the number of boat trips) should be maintained by site, date, and time period, and spatiotemporal sampling units should be defined for a stratified PPS sampling approach. For any given selected combination of landing site and time interval, it will be necessary to obtain accurate counts of all boat trips and either census or conduct probability sampling of those boat trips. Ideally, samplers would be able to obtain a copy of the captain’s logbook for the trip prior to conduct-ing interviews with anglers who fished on the boat. The sampler should attempt to interview as many anglers who fished on the boat as possible, but if sampling must be random if not all anglers can be interviewed. It is also very important to employ


probability sampling techniques to get measurements on a representative subset of the fish in each angler’s catch.

For headboat fishing, it is desirable to conduct an independent validation survey that would send observers out on a probability sample of headboat fishing trips so they can directly observe at least a subsample of the fish that are caught by anglers and released alive. For this purpose, a boat list frame with expected fishing activity in-formation on all boats could be used to perform PPS sampling of boats. The boats with more expected fishing trips should be given a greater likelihood of being se-lected than the less expected activity. Once a given boat is selected, a random sam-pling of scheduled fishing trips for that boat would determine which trip to sample. Ideally, the sampler would be able to directly observe and record all of the landed catch of anglers who fished on the boat, but would only observe a randomly selected subset of the anglers to observe and measure their released catch. Probability sam-pling should be employed to get measurements on a subset of the fish landed by each angler.

Estimators of catch for charter boats or headboats should be based on data collected in both the logbooks and the parallel sampling surveys. Combined estimators could be developed that would allow for correction of any consistent self-reporting errors in the logbook data provided by the boat operators. The data obtained for sampled anglers serves as a check against the data obtained from the boat operators. Logbook surveys conducted without the necessary sampling for validation are likely to pro-vide erroneous results.

In 2010, the MRIP initiated a pilot study in the Gulf of Mexico to test the use of a “logbook-access” design for monitoring catch and effort in the charter boat fishery. Operators of federally permitted charter boats in Texas and in the panhandle region of Florida were asked to submit logbook data on each of their fishing trips through the use of available electronic reporting technologies. In addition, an access point survey was used to intercept boat trips and collect catch data from individual anglers. This study was intended to test a “best practices” design and results will be evaluated to assess the feasibility of such a design for broader application to monitoring fishing on both state and federally permitted charter boats in the Gulf of Mexico. Preliminary results indicate that compliance has been an issue, with operators of only about 70-80% of the Florida Panhandle boats submitting their reports on a regular basis. Data collection has been completed, and a final report of results is expected in early 2012


Annex 1: List of participants

Name Address Phone/Fax Email

Anssi Ahvonen PO Box 2 00791 Helsinki Finland

Phone +358 405 222219 [email protected]

Trude Borch

Norut - Northern Research Institute Norut Tromsø Postboks 6434 9294 Tromsø Norway

Phone: Fax:

[email protected]

Miguel Cabanellas IMEDEA Spain

[email protected]

Frederico Cardona Pons

IMEDEA Spain

[email protected]

Håkan Carlstrand Swedish Board of Fisheries P.O. Box 423 SE-401 26 Gothenburg Sweden

Phone: Fax:

[email protected]

Martin De Graaf Wageningen IMARES P.O. Box 68 NL-1970 AB IJmuiden Netherlands

Phone: Fax:

[email protected]

Tessa van der Hammen

Wageningen IMARES P.O. Box 68 1970 AB Ijmuiden Netherlands

[email protected]

Kieran Hyder

Centre for Environment, Fisheries and Aquaculture Science (CEFAS) Pakefield Road NR33 0HT Lowestoft Suffolk United Kingdom

Phone +44 1502 524501

[email protected]

Belle-Rosa Khoung

FOD Public Health, Food Security and Environment Eurostation II, Place Victor Hortaplein, 40 bus 10 1060 Saint Gilles Belgium

Phone +32 25249797

[email protected]

Harold Levrel Chair

IFREMER Centre de Brest Marine Economics Department Technopole Brest-Iroise BP 70 29280 Plouzané France

Phone: +33 2 29 00 85 27 Fax: +33 298224776

[email protected]

Anne McLay

Marine Scotland Science Marine Laboratory P.O. Box 101 AB11 9DB Aberdeen United Kingdom

[email protected]

mailto:[email protected]





Name Address Phone/Fax Email Beatriz Morales-Nin

Mediterranean Institute for Advanced Studies C/ Miquel Marquès, 21 07190 Esporles, Mallorca, Iles Balears, Spain

Phone +34 971611721 Fax +34 971611761

[email protected]

Estanis Mugerza AZTI-Tecnalia AZTI Sukarrieta Txatxarramendi ugartea z/g E-48395 Sukarrieta (Bizkaia) Spain

Phone +34 94 6029446 Fax:

[email protected]

Vidar Öresland

Swedish Board of Fisheries Institute of Marine Research, Lysekil P.O. Box 4 453 21 Lysekil Sweden

[email protected]

Miquel Palmer Vidal

Mediterranean Institute for Advanced Studies C/ Miquel Marquès, 21 07190 Esporles, Mallorca, Iles Balears, Spain

[email protected]

Krzysztof Radtke

Sea Fisheries Institute in Gdynia ul. Kollataja 1 PL-81-332 Gdynia Poland

Phone +48 587356269 Fax +48 58 7356 110

[email protected]

Hanne Rasch

Directorate of Fisheries PO Box 185 NO-5804 Bergen Norway

Phone +47 92665796

[email protected]

Claus Reedtz Sparrevohn

The National Institute of Aquatic Resources Jægersborgvej 64–66 DK-2800 Kgs. Lyngby Denmark

Phone: Fax:

[email protected]

Delphine Rocklin IFREMER Centre de Brest Marine Economics Department Technopole Brest-Iroise BP 70 29280 Plouzané France

Phone: +33 2 98 22 45 85 Fax: +33 298224776

[email protected]

Harry V. Strehlow Johann Heinrich von Thünen-Institute, Federal Research Institute for Rural Areas, Forestry and Fisheries, Institute for Baltic Sea Fisheries Alter Hafen Süd 2 18069 Rostock Germany

Phone: +49 3818116107 Fax:

[email protected]







Name Address Phone/Fax Email Remko Verspui

Royal Dutch Anglers Association n/a n/a Netherlands

Phone +31 30 6058400

[email protected]

Dave Van Voorhees Chair

National Marine Fisheries Services 1315 East West Highway Silver Spring MD 20910 United States

Phone 1-301-713-2328 Fax 1-301-713-1875

[email protected]

Jon Helge Vølstad Institute of Marine Research P.O. Box 1870 Nordnes 5817 Bergen Norway

Phone: +47 55 23844 Fax:

[email protected]

Lucia Zarauz

AZTI-Tecnalia AZTI Sukarrieta Txatxarramendi ugartea z/g E-48395 Sukarrieta (Bizkaia) Spain

Phone +34-946029400

[email protected]



Documents

ICES PGRFS REPORT 2011 Reports/Expert Group...Denmark Telephone (+45) 33 38 67 00 Telefax (+45) 33 93 42 15 [email protected] Recommended format for purposes of citation: ICES. 2011. Report