Advisory Committee on Fishery Management ICES CM 2003/ACFM:06

REPORT OF THE

ICES/EIFAC Working Group on Eels

Nantes, France 2–6 September 2002

This report is not to be quoted without prior consultation with the General 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.

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

Palægade 2–4 DK–1261 Copenhagen K Denmark TABLE OF CONTENTS Section Page

1 INTRODUCTION...... 1 2 TRENDS IN RECRUITMENT, FISHING YIELD, SOCIO-ECONOMICAL IMPLICATIONS...... 2 2.1 Trends in recruitment...... 2 2.2 Trends in stock and Yield ...... 2 2.3 Trends in re-stocking ...... 3 2.4 Trends in aquaculture...... 5 2.5 Socio-economic data trends ...... 6 2.5.1 France ...... 6 2.5.2 Netherlands ...... 7 2.6 Improvement of the database for assessment of the European eel stock...... 7 3 IMPACT OF EXPLOITATION ON SPAWNER ESCAPEMENT...... 9 3.1 Introduction...... 9 3.2 Exploitation of glass eel...... 9 3.2.1 Adour ...... 9 3.2.2 Vilaine...... 10 3.2.2.1 Impact of glass eel exploitation on spawner production ...... 11 3.3 Exploitation of yellow eel...... 11 3.4 Impact of silver eel fisheries ...... 12 3.5 Conclusion on impact of exploitation on spawner escapement...... 14 4 DENSITY DEPENDENT PROCESSES IN THE CONTINENTAL LIFE STAGES ...... 15 4.1 Introduction...... 15 4.2 Migration in relation to stock density ...... 15 4.3 Sex ratio and stock density...... 17 4.4 Mortality ...... 18 4.5 Growth ...... 18 4.6 Conclusion on density dependent processes ...... 19 5 HABITAT LOSS ...... 20 5.1 Introduction...... 20 5.2 Dams ...... 20 5.3 Hydropower stations and pumping stations ...... 21 5.3.1 Damage and mortality at turbines and pumping stations ...... 21 5.3.2 Damage and mortality by passage through the turbine...... 21 5.3.3 Damage and mortality by passage through a pump ...... 22 5.3.4 Damages in the tailwater...... 22 5.3.5 Damages before the turbine and protection devices...... 22 5.3.6 Cumulative mortality ...... 22 5.3.7 Hydropower generation and capacity: status, prospects and impacts on the European eel stock ... 23 5.4 Habitat loss ...... 23 5.5 Case studies...... 24 5.5.1 Villaine ...... 24 5.5.2 Garonne-Dordogne-Charente...... 24 5.5.3 The Frémur ...... 26 5.5.4 The Meuse...... 28 5.5.5 The Rhine...... 29 5.5.6 North America ...... 30 5.6 Conclusions on habitat loss...... 31 6 REFERENCE CONDITIONS, EXPLOITATION AND HABITAT AVAILABILITY...... 32 6.1 Introduction...... 32 6.2 Reference conditions for northern type stocks...... 32 6.3 Southern type fisheries and stocks of growing eel...... 34 6.4 Possible levels of habitat restoration targets ...... 34 6.5 Use of stock density measures in definition of reference conditions ...... 34 6.6 Conclusion on reference conditions...... 35 7 EXPLOITATION VS. OTHER MEASURES ...... 36 7.1 Introduction...... 36 7.2 Turbine mortality ...... 36 7.3 Habitat accessibility ...... 36

i Section Page

7.4 Predation ...... 37 7.5 Restocking ...... 37 7.6 Conclusions on exploitation vs. other measures ...... 38 8 POST-EVALUATION - DEVELOP DATA-RICH AND DATA-SPARSE PROCEDURES FOR EVALUATING THE EFFICACY OF MANAGEMENT MEASURES...... 39 8.1 Introduction...... 39 8.2 Development of Procedures...... 39 8.3 Timeliness...... 39 8.4 Feasibility of detecting a change in the indicator, and its implications...... 40 8.5 Methods for collecting and analysing relevant data...... 40 8.5.1 Glass Eel ...... 41 8.5.2 Yellow Eel ...... 41 8.5.3 Silver Eel ...... 41 8.6 Conclusions on post-evaluation ...... 42 9 CONTAMINATION AND EELS...... 43 9.1 Introduction...... 43 9.2 Status...... 43 9.3 Legislation ...... 43 9.4 Conclusions on contamination ...... 44 10 CONCLUSIONS AND RECOMMENDATIONS...... 45 10.1 Conclusions...... 45 10.2 Recommendations...... 45 11 LITERATURE REFERENCES ...... 46 APPENDIX 1. LIST OF PARTICIPANTS...... 62 APPENDIX 2: OVERVIEW OF HABITAT CHANGES PER COUNTRY...... 64 @#

ii

1 INTRODUCTION

At the 88th Statutory Meeting of ICES (2001) and at the 22nd meeting of the EIFAC in Windermere, United Kingdom, it was decided that:

The ICES/EIFAC Working Group on Eels [WGEEL] (Chair: W. Dekker, Netherlands) will meet in Nantes, France from 2–6 September 2002 to: a) identify the priority list of stocks for which assessment information will be collated and analysed; assess the trends in their recruitment, stock biomass and yield; and analyse their causes; b) assess the impact of eel fisheries on local populations by using criteria (or their proxies) developed for data-poor situations, and determine whether this impact compromises existing escapement targets; c) assess whether growth, mortality and migration are density-dependent, and if so how this could affect the production of spawners; d) assess the type and extent of habitat loss by river system, region, and country, and derive targets for habitat restoration to achieve appropriate biological goals for eel stocks; e) compile handbooks to help managers:

i) describe the characteristics of an unexploited eel stock for use in setting management objectives (this may require analytical studies or empirical comparison between exploited and unexploited stocks),

ii) quantify the effects and risks associated with exploitation and loss of habitat, and the corresponding effects of ameliorating management measures such as habitat restoration, the construction of fish passes, or re-stocking,

iii) develop data-rich and data-sparse procedures for evaluating the efficacy of management measures,

iv) assess the effect of fishing on the economic viability of local communities and management.

24 people attended the meeting from 11 countries (see appendix 1).

The current Terms of Reference and Report very much constitute one step in an ongoing process of documenting the eel stock and fisheries and compiling management advice. As such, the current Report does not present a comprehensive overview, but should be read in conjunction with previous reports (ICES 2000, 2002).

During the meeting of the Working Group, it was felt that ongoing management and research of eel necessitated consideration of some issues that were not fully included in the Terms of Reference. It was decided not to exclude these items from the discussions and consequently this report also contains some discussion not directly related to the TORs. This applies in particular to the presentation of indicator data series on the status of the stock and fisheries, human food safety and pollution in eel and its possible relation to stock dynamics.

The structure of the report essentially follows the Terms of Reference for the meeting, with some additional sections added where appropriate.

1 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

2 TRENDS IN RECRUITMENT, FISHING YIELD, SOCIO-ECONOMICAL IMPLICATIONS

2.1 Trends in recruitment

There are relatively few data sets which provide information on the recruitment of the European eel and the information they provide relates to various stages (pigmentation, behaviour) of the recruitment into continental habitats (Dekker 2002). Available time-series from 19 river catchments in 12 countries were examined for trends (Table 2.1.1). The data analysed were derived from both fishery-dependent sources (i.e. catch records) and fishery-independent surveys across much of the geographic range of the European eel, and cover varying time intervals.

Over the last two decades of all time-series, downward trends were evident, reflecting the rapid decrease after the high levels of the 1970s. Over the 1980s, the trend was downwards with the exception of the Erne in northwestern Ireland in which no trend was apparent. In the 1990’s most series have shown fairly stable low levels, while 2001 resulted in a historical minimum. Most recent data show a continued decrease and the 2002 level is not showing a substantial recovery from the 2001 historical minimum for the whole stock. For the trap in the Göta Älv (Sweden), operation has been re-continued in 2002, following a 4 years closure, resulting in an exceptionally high catch.

10000

1000

100 lue

% of 1979-1994 va 10

1

Göta Älv Viskan Bann Erne Shannon DenOever Ijzer Vilaine (total catch) Tiber 0 1950 1960 1970 1980 1990 2000

Figure 2.1.1 Time-series of glass eel monitoring in European rivers, for which data are reported in 2002. Each series has been scaled to the 1979-1994 average. The 2002 data for Göta Älv (Sweden) are influenced by discontinued operation in the previous 4 years.

2.2 Trends in stock and Yield

The Food and Agricultural Organisation FAO (Rome, Italy) of the United Nations maintains a database of fishing yields. Additionally, the International Counsel for the Exploration of the Sea ICES (Copenhagen, Denmark) maintains a database of landings of marine, Atlantic fishing yields. Since the data in the ICES database exclude the major yield from the stock at forehand, preference was given to the FAO data.

Official landings statistics for many countries comprise only about half of the true catches in the 1980s and 1990s (ICES 1988; Moriarty & Dekker 1997), because of illegal and unreported catches, as well as lack of coverage of many areas in several countries. However, to some extent trends in the reported data will reflect true changes in fishing yields.

FAO eel landings statistics (with minor corrections) are presented in Table 2.2.1 and Figure 2.2.1. The data show a clear decrease of yield for all countries in recent years.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 2

The FAO catch return data do not necessarily reflect the status of the eel stock. Effort can be variable and underreporting the catches is a serious problem in most countries. Additionally, some data reported for outdoor fishing clearly include indoor aquaculture yield. This has been the case in Denmark, Netherlands and Italy. Where available, this has been corrected.

20,000

18,000

16,000 Norway 14,000 Sweden Denmark Germany 12,000 Ireland UK Netherlands 10,000 France Spain

Landings (tonnes) 8,000 Portugal Italy Remaining Europe 6,000 Northern Africa

4,000

2,000

0 1950 1960 1970 1980 1990 2000

Figure 2.2.1 Landing statistics of the European eel in the past 50 years.

FAO data, with minor corrections. Last years presumably incomplete.

2.3 Trends in re-stocking

Data were obtained from a number of countries, separate for glass eels and for young yellow eels. The size of ‘young yellow eel’ varies between countries. Most data available were on a weight base. Weights were converted to numbers, using estimates of average individual weights of the eels stocked. These were 3.5 g for Denmark, 33 g for the Netherlands, 20 g for (eastern) Germany, and 50 g for Sweden. An overall number of 3000 glass eels per kg was applied.

Recent time-series available were available from (eastern) Germany, Netherlands, Sweden, Denmark, Northern Ireland and Belgium (Flanders). For Poland, an older time-series was available. These are presented in Tables 2.3.1.1 and 2.3.1.2.

Trends in the levels of re-stocking for 2001 and 2002 cannot be identified due to a lack of data. The only data available shows a small increase in glass eel stocking in the Netherlands and a small fall of young yellow eel stocking in Germany.

3 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

140

120 D east NL S Dk Flanders N.Irl. PO

100 eels) f o s n 80 illio m ls ( e 60 lev g n i k c o t 40 S

20

0 1950 1960 1970 1980 1990 2000

Figure 2.3.1.1 Re-stockings of glass eels (data available from individual countries; some relevant countries are missing). Last years presumably incomplete.

12

10 D east NL S Dk Flanders eels) 8 illions of

m 6

4 ocking levels ( t S 2

0 1950 1960 1970 1980 1990 2000

Figure 2.3.1.2 Re-stockings of young yellow eels (data available from individual countries; some relevant countries are missing). Last years presumably incomplete.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 4

140

D east NL S Dk N.Irl. Flanders PO 120

100 eels) 80 illions of m

60 Stocking levels (

40

20

0 1950 1960 1970 1980 1990 2000

Figure 2.3.1.3 Re-stockings of glass and young yellow eels, numbers re-stocked (data available from individual countries; some relevant countries are missing). Last years presumably incomplete.

140

120

yellow eel glass eel 100

80

60 Stocking levels (millions of eels)

40

20

0 1950 1960 1970 1980 1990 2000

Figure 2.3.1.4. Re-stocking of glass eel and young yellow eel, by life stage.

There is no information on re-stockings available for the Western Germany and other central European countries, where re-stocking is known to take place.

2.4 Trends in aquaculture

Aquaculture of the European eel ranges from highly industrialised, indoor facilities in northern Europe, through extensive culture in artificial ponds in southern Europe, to re-stocking of foreign glass eel in semi-natural outdoor

5 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

waters for fisheries in northern Europe. All aquaculture fully depends on seed stock derived from the wild population, since artificial reproduction fails in the young larval stage. Additionally, aquaculture plants are used for quarantine of foreign glass eel to be re-stocked in outdoor waters (e.g. Sweden, Germany) and transports of partly grown eels in- between aquaculture and fisheries occurs in and between countries (France, Italy). Obviously, the distinction line between aquaculture and fisheries is hard to define.

For aquaculture production, no consistent long running time-series exists. Data are available from FAO, from the Federation of European Aquaculture Producers, from previous meetings of the working group and from Kamstra (1999). An overview of the estimates is compiled in table 2.4.1

The aquaculture production in Europe is concentrated in Denmark, the Netherlands and Italy. The aquaculture in Denmark and the Netherlands is technically speaking highly developed and produces an increasing part of the total, while Italy has intensive as well as extensive culture systems, the latter with a declining production. Lack of sufficient data for 2001 aquaculture production has resulted in little possibility of identifying any changes in trends.

12000

others/unident. EU Italy 10000 Netherlands Denmark

8000

6000

4000

2000

0 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001

Figure 2.4.1. Trends in aquaculture production of the European eel. Last years for some of the other countries presumably incomplete.

The aquaculture industry is growing all around Europe for several species such as salmon, trout, carp, sea bass, sea bream, oyster, mussels, turbot and eel. In all cases but the eel, the unit price has fallen between 1988 and 1997. In this time, the production of eel has increased by 33% and its unit price by 24% (European Commission Fisheries Directorate General: Forward Study of Community Aquaculture, MacAlister Elliott and Parners Ltd 1999).

2.5 Socio-economic data trends

2.5.1 France

Figure 2.5.1.1 shows the contribution eel fishery, and in particular glass eel fishery makes to the overall yield from the Gironde system (Castelnaud et al 2001). The financial contribution of glass eel has risen from 19% in 1978 up to a peak in 1997 of 77%. The Gironde fisheries rely on glass eels for a large percentage of their income, while the glass eel in itself is experiencing falling catches. The high share of eel in the total value is the result of a rise in unit price, faster than the decline in catch.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 6

others 2500 100 others yellow eel yellow eel glass eel glass eel 2000 75 ) s 1500 FF)

(M 50 e h (tonne lu tc 1000 a Va C 25 500

0 0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998

Figure 2.5.1.1 Fishing yield in Gironde (France) in weight and value. Note the increased contribution of glass eel.

2.5.2 Netherlands

The local economical value of an eel fishery can be important since rural development can often be reliant on the eel market (Figure 2.5.2.1). In lake IJsselmeer (Netherlands) it is apparent that there is a decline in the number of fishermen (number of companies/licences) coupled with a fall in yellow eel landings over the last 30 years. However, the number of fishing gears has significantly increased from 1970 to 1988, fyke net numbers rising 300% in 18 years, leading to the conclusion that although licences numbers have decreased the number of nets/boxes per fisherman has risen in response to the falling stock levels. As a consequence, the local stock is extremely overexploited (Dekker 2000c)

5000 60000

4500

50000 Fykenets 4000

3500 Landings 40000

3000 Number of fyke nets

2500 30000

2000

20000

Number of licenses / landings (tonnes) 1500

1000 10000 # licenses 500

0 0 1950 1960 1970 1980 1990 2000

Figure 2.5.2.1 Eel fishery in Lake IJsselmeer. Fishing effort expressed by number of fykes nets and eel boxes vs. the total landing and the number of companies in 1987-1999 period. (Data from Dekker 1991).

7 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

2.6 Improvement of the database for assessment of the European eel stock

There is currently no formal reporting procedure listing the national data required for annual update to the Working Group on Eels. Enhanced and structured data provision would dramatically improve annual progress at the Working Group meetings. A standing request to national participants could be made for data on:

• Landings of glass, yellow and silver eels, subdivided quantitatively into catches of different gears and with catch per unit effort data • Numbers of fishermen fishing or number of licensees • Data on prices for each life stage • Details of scientific (fishery independent) surveys of stocks • Relevant environmental information (habitat change, water quality, bioaccumulation of contaminants) • Number and of hydropower dams and their characteristics.

While some of these data will be subject to little year on year change, others (particularly the fishery data) will have significant interannual variation. Some of these data may not necessarily be addressed at each annual meeting, but be assessed at periods relevant to their rate of change

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 8

3 IMPACT OF EXPLOITATION ON SPAWNER ESCAPEMENT

3.1 Introduction

The maturing stages of eel have never been observed in the wild, but are undoubtedly purely oceanic in nature. Escapement of silver eel from the continent provides the best indicator of oceanic spawning stock biomass, but silver eel escaping the continental fisheries are probably more correctly defined as pre-spawners. There are no means available to assess potential losses between silver eel emigrating from freshwater and the oceanic spawning phase in the life cycle. Consequently, discussion will focus on the impact of fisheries on silver eel escapement.

The information available with regard to the impact of fisheries on silver eel escapement is very limited relative to the number of fisheries operating and there are few estimates of fishing mortality. It is impossible to assess the effect of fisheries on the overall escapement of the European eel stock with any real confidence (Dekker 2000a&b) as there are insufficient data and existing estimates for specific fisheries are mostly rather crude. Hence, stock-wide management targets can not be derived. However, the available information indicates that fisheries on all life-stages can and often do impact upon spawner escapement within particular locations and further suggests that some fisheries are capable of completely precluding escapement of potential spawners from a catchment or fishery. It follows that further controls on local fisheries on all components of the stock are appropriate and should contribute to the overall enhancement of production and escapement of spawners.

Most systems have insufficient data on which to set escapement targets now or in the foreseeable future. Only the few data rich systems can allow ready implementation of any limit reference value. In the long-term, mathematical models may be developed for the total stock based on individual monitoring of its components, and the methods chosen now to set limits should encourage the collection of the necessary data and to derive proximate criteria for data poor environments. In the short-term, however, priority must be given to those systems for which available information is adequate to implement rational management. For all other systems, with less than ideal data conditions, some approximate management scheme must be adopted, based on knowledge derived from data rich systems. This chapter discusses the selection of data rich cases, for which the impact of fisheries on spawner escapement is known, or can be derived in the short-term.

3.2 Exploitation of glass eel

There are two case studies available on the effect of exploitation on the local glass eel stock: the glass eel fisheries in the Adour (France) and the Vilaine (France).

3.2.1 Adour

On the Adour estuary observations have been made on the behaviour of glass eel during their upstream migration into the lower part of the estuary in the framework of an EU program (DG XIV – 99/025).

Densities and biomass have been estimated on fixed stations from an adapted sampling design during 3 fishing seasons (1998-2000). 31 daily scientific surveys have been made during this period. Professional catches, in the maritime fishery using two push-nets, located below the sampling stations have been recorded in parallel so some daily rates of exploitation of the glass-eel runs have been estimated.

The Table 3.2.1 below summarised the different estimations made during the 3 fishing seasons.

Table 3.2.1: Exploitation rates in the Adour estuary Estimated daily rate of 1998/1999 fishing season 1999/2000 fishing season 2000/2001 fishing exploitation season Mean estimate 16% 13% 30% Median estimate 6% 9% 26%

The rate of exploitation is highly variable with a 20-fold variation between the minimum and the maximum. Studies on the effect of hydroclimatic factors on catchability showed that the main factor controlling the accessibility of glass eels is the clarity of the water column: low turbidity is always associated with a low density of glass eel close to the surface and consequently a low catchability. Push-nets sample only a layer of two meters from the water surface.

9 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

3.2.2 Vilaine

The Vilaine watershed covers 10 400 km2; 30% of Brittany (NW of France). 12 km from the sea, a dam closes the estuary. Hydraulic conditions, near the gates of the dam prevent glass eel from moving upstream.

A commercial glass eel fishery is concentrated within 2 km from the dam. Fishermen use two 1,20 m diameter nets-of 2 mm mesh size-pushed on each side of the boat. The nets may be placed either on the surface or deep. Accurate statistics have been collected since 1996 by a method allowing the selection of a reliable sample and the extrapolation to the whole fishery. At the same time, the glass eel migration towards freshwater has been monitored on a glass eel trap located on the dam. (Briand, Fatin, Feunteun & Fontenelle In press).

The analysis of glass eel biology and exploitation shows that the glass eel fishery is very intensive and that there is probably 100% exploitation during the fishing season. The escapement for the entire from the fishery is estimated to range between 1 to 10%. In such a fishery, the fishing effort affects the abundance and total catches have to be used instead of CPUE as an estimation of abundance.

From 1998, fisheries regulations have been introduced to limit the extension of the fishing season. The length of season was reduced from 30 April to 23 March in 2002. These regulations were aimed at ensuring the escapement towards freshwater of at least 700 kg of glass eel per year. Because of the continuous decrease in stock from 1996 to 2001 and a very low April recruitment they only attained the level of 700 kg in 1998 (corresponding to 2 tons of escapement in estuary given a fish-pass efficiency of 30%), and levels of 70 to 450 kg in other years.

In 2002, new regulations were also applied: the efficiency measured for the glass eel fish pass was calculated to be only 30 % of the total estuarine stock arriving after the fishing season (Briand, Fatin, Fontenelle & Feunteun, submitted). An analysis of the otoliths shows that those springtime recruits arriving after the fishing season and remaining in the estuary make little contribution to the yellow eel stock. More than one ton of glass eel remaining in estuary was estimated to contribute only to 15 % of the estuarine population in the Vilaine for the 1998 cohort. The other part was made of glass eel arrived in autumn and having settled before the starting of the fishing season. This contradiction may be demonstrating the existence of a density dependent mortality, which seems to have been particularly important in springtime for glass eel whose migration was inhibited by the dam. It was thus decided to introduce new regulations and to catch glass eel in estuarine water while the fishery was closed to ensure the migration towards freshwater of at least 60 % of the stock.

One of the main outputs of the Vilaine study is that a low escapement has led to a substantial increase in the yellow eel density measured by electrofishing. But a large uncertainty remains on the effect of the measures with regard to the stock. The escapement of 700 kg set as a target by regulations remains questionable.

Table 3.2: Presentation of the selected cases Site Description Fishery parameters Escapement from the fishery Vilaine 1996-2002 The fishery is located below the dam. No escapement 1-10% of the incoming The number of fishermen has during the fishing recruitment fluctuated from 90-150 fishermen over season the past 8 years. Fishing gear used: Yield = 16 T in 2002 Push nets with 1.20 m diameter Effort = 15,300- Regulation: Fishery is open from 15 35,000 h November to 23 March with Sundays and holiday's closures. Fishery is open 3,400-7,300 trips daily from 6 pm to 8 am. CPUE 2.75-1.2 Adour 2000 The fishery is located in estuary and 60 fishermen Daily escapement 70-87% tidal zone of the estuary and is Yield = 2 t push net of the incoming recruitment composed of a scoop net and push net Effort 2100 trips (2 nets / boat) fishery Calculations of 39 trips per fisherman escapement rate only apply to the push CPUE push net = 0.9 net fishery kg/trip in 2001 but The 2-push net fishery is open from 15 ranged from 0.7 to November to 31 March with Sundays 4.1 kg/ trip 1984- and holidays closures. 2000

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 10

3.2.3 Impact of glass eel exploitation on spawner production

There is no single comprehensive study on the impact of glass eel exploitation on the net spawner production in any river system available. The impact of glass eel exploitation on the local population of yellow eel is largely unknown too. The fisheries of the Vilaine and the Adour estuaries are probably an indication of the range of efficiency of glass eel fisheries. The Adour fishery is located in the downstream part of the estuary, where glass eel only remain for a short while before being carried upstream by the tide. It provides an idea of the efficiency of an estuarine glass eel fisheries. The case study of the Vilaine illustrates that, in an area where glass eel concentration occurs, the fishing efficiency may be very high. Several French estuaries are closed and present the same context than the Vilaine (Somme, Douve, Vire, Baie de Bourgneuf, Sèvre Niortaise, and Charente). As the behaviour of glass eels leads them to gather near a dam, the fishery efficiency may be high enough to deplete local stock.

But the major fisheries (Loire, Gironde) are located in opened estuaries. Even without dam to impede the glass eel migration, a concentration of glass eel will occur eel at the tidal head of these estuaries, because of the mechanism of selective tidal stream transport (McCleave et Wippelhauser 1987, Gascuel 1986, Gascuel et al. 1995). Escapement from the places where the fisheries concentrate their effort is unknown. It will depend on the delay necessary to observe either counter current migration or settlement. Active migration are never observed when temperature are lower than 8°C (Deelder 1952, 1958; Hvisdten 1985, Gascuel 1986; McGovern et McCarthy 1991) and a temperature close to eleven degrees induces a reduction in catchability on the Loire (Desaunay et al. 1987).

There is probably only a short penetration of glass eel stage observed in freshwater streams. Glass eel are found on weirs at the tidal head of the estuary when active migration allows them to progress within the estuary (Gascuel 1986, McGovern and McCarthy 1992). Nevertheless, they are rarely encountered numerously far upstream from the estuary limit. 0+ elvers are scarce at 15 km from the estuary on the river Shannon (Moriarty 1986). Identically, elvers only penetrate short distance upstream brackish waters (Naismith and Knights 1988; Lobón-Cerviá et al. 1995).

It is assumed that in-between the immigrating glass eel stage and the settling yellow eel phase, a large and probably density related mortality occurs (see chapter 4). However, there are very few data on mortality in this stage. Mortality during estuarine, passive migration is also thought to be high (Knights, Bark, Ball, Winter & Dunn 2001). During an experimental survey in the Aulne estuary in 2000 and 2001, mortality of a sub-sample collected and placed in aquaria was low (0.5%) when the sample was collected low in the estuary, but just below the tidal limit, the mortality increased to 100 % (Briand 2001).

3.3 Exploitation of yellow eel

Yellow eel are exploited in a wide range of habitats throughout Europe using a range of fishing gear types. Analyses of both yellow and silver eel catches from many countries are complicated by the fact that most of the information available relates to the combined silver and yellow eel landings.

Overall, yellow eel dominates the landings in weight and is exposed to high levels of exploitation because of its exposure to fishing for a long time during its life stage in freshwater. The time spent in freshwater explains the discrepancy in the escapement rates between males and females as reported by Dekker (2000c) in IJsselmeer (Table 3.3).

There are many analyses of the impact of fisheries on a local eel stock. However, there are just two studies on the net effect on spawner escapement (Svedäng 1999; Dekker 2000c). In contrast to the Swedish East and South coast, the West Coast fisheries focus on yellow eel. Assuming neither recruitment nor total mortality has changed in time, an estimate of (total and fishing) mortality can be derived from length frequency data (Ricker curve). On the Swedish West Coast, instantaneous fishing mortality is estimated at F=0.31. This conforms to a reduction in female spawner escapement to 15 % of the unexploited state, over the 6.2 years in-between minimum legal size and average size at maturation of females. Lake IJsselmeer (Netherlands) has an extremely overexploited yellow eel fishery, in which only one out of seven males escapes exploitation, while for females, only one in seven hundred escapes. This extremely low escapement rate is directly related to the fishing mortality in the yellow eel stage, which is in the order of F=1.0 in lake IJsselmeer. Simulation of (higher or) lower fishing pressure shows, an F=0.2 would reduce female spawner production to 30 % of the (simulated) unexploited state.

These results can not be extrapolated to other settings without consideration of differences in exploitation pattern, growth rate and possibly migratory behaviour. However, a review of case studies (table 3.3) shows considerable variation in fishing intensity between areas, which probably also indicates considerable differences in the impact on spawner escapement.

11 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

A decline in the silver eel catches in some areas may be a result of increased fishing for yellow eel. In Lough Neagh (Northern Ireland), an increased fishing effort for yellow eels, initiated in order to increase employment on the fishery, led to a decline in the silver eel output from this system (Rosell 2000, 2001).

Table 3.3 Overview of studies assessing the impact of yellow eel fisheries on the local stock. Site Description Fishery parameters Fishery Impact Lake IJsselmeer, The area open to fishing in Average production is The eel stock of the Netherlands Ijsselmeer lake = 1,820 about 165 kg/ha.N total Ijsselmeer is heavily 1989-1996 Km2.Fishing gear used : population =23 millions in overexploited Escapement Dekker (2000c) Fyke nets, eel boxes, and 1989, 8 millions in 1996. rate is 1/7 for males and longlines.Number of Natural mortality 1/700 females of a virgin fishermen = 65 companies (M=0.138). stock. Production is estimated at 300 T/year Shannon River1992-2000 Fishing gear used is fyke CPUE declining in the The eel stock in the lower nets and longlinesThe lower catchment from 0.5 portion of the Shannon number of licensed fishing to less than 0.1 River is considered at crew is about 40 (2 kg/trap/night. medium levels of fishermen per crew).The exploitation , while the total catch ranged from 10 upper portion seems is in to 50 T.Each crew is fair condition. allowed up to 50 nets or 1000 hooks.Fishing season is open from May to September, weekdays only. Grand Lieu Lake (Loire Fishing gear : 3 trousers Growth rate (G=36.65 mm The eel stock of the Grand River)1990-1999 deep nets, bottom hook and (males) à 62.45 mm Lieu lake is heavily line, and eel pots. (females). Total mortality exploited with minimum Adam (1997) Fishermen are allowed 10 (Z=0.91 à 0.972). Natural escapement. fyke nets (@ 10 mm mesh mortality is estimated at size) Average yield is M=0.2 +/-0.1. estimated at about 25 T Swedish West Coast 1993- Annual yield about 25-50 T CPUE = 0.2-0.5 kg/fyke Catches are stable while 2000 a year.Area fished is about nets/ day.Total mortality is recruitment has decreased 8,600 km2 with depths less estimated between 0.31 to in recent years. Escapement Svedäng (1999) than 20 meters. 0.54.Natural mortality = is estimated at 0.2 0.23. Mortalités moyennes million/year. totale Growth rate (G=45 mm for females from 4 to 11 years). The stock size is estimated at 11 million eels/year (>370 mm). Recruitment is estimated at 5.4 million glass eels/year. The Corrib lake system, Fishing area is 26,500 CPUE is estimated at 0.8- The yellow eel fishery in Ireland Moriarty 2002. haFishing gear used is fyke 1.2 kgs/fyke net/night. the Corrib lake systems nets and longlines.Fishing Growth rate is estimated at seems to have little to no

licenses = 18 2.3 cm/year. effect on the silver eel Yield for 2001 = 8,949 kgs fishery. Escapement rate has been estimated at 40- 50%.

3.4 Impact of silver eel fisheries

With increasing northern latitude in Europe the proportion of silver eel catches tend to increase, with Scandinavian countries producing highest proportions. There is no accurate information on the percentage of silver eel escaping from the silver eel fishery.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 12

Mark recapture studies have been conducted in very large (Baltic) systems and some bigger rivers. Escapement rates of silver eel range from 60-90%.

For the population as a whole, the impact of silver eel fisheries might be smaller than in these individual case studies. The extent of silver eel fisheries in Europe is generally limited to the larger river systems and the mark-recapture experiments have all been confined to larger water systems. Other, mostly smaller systems do contribute to the overall spawner production, while exploitation of the silver eel stage is generally rare. Consequently, the locally estimated impact of silver eel fisheries only sets an upper bound on the overall impact. There is currently no continent wide overview of river systems and eel exploitation, detailed enough to assess the overall impact.

Table 3.4 Overview of studies assessing the impact of silver eel fisheries on escapement. Site Description Fishery parameters Impact Escapement from the fishery Loire River The fishery is located from 80 Yield : 50 T One mark-recapture gives an Feunteun & to 300 km upstream to the escapement of 80-90% of the Effort : 80 nights per Boisneau, estuary migrating eels. fishermen, 960 nights for Number of fishermen: 12. the all fisheries and hall Fishing gear used : guideau CPUE : 0-500 fishes per (individual coghill) night Regulation : Fishery is open from mid September to end February day and night. Exploitation stops in the end of December. Baltic sea The fishery is located in lakes Yield : 1,000 T Long-term mark-recapture Moriarty (1997) and coastal brackish waters of experiments give a mean Ask and Erichsen southern and eastern parts of escapement of 60% of the migrating (1976) Sweden. eels. Sers et al. (1993) Number of fishermen : 356 Westin (1990) fulltime 562 part-time and 399 Pedersen & occasional Dieperink (2000) Fishing gear used : pound net

Regulation : fishing is allowed all year long but effective only from May to November Erne River One fishery is located in Lower Yield : 10-15 T Mark-recapture experiments in Erne another in Cavan 1998 and 1999 give a mean Matthews et al. Effort in the river : not Monagahan lake escapement of 81-84% of the (2001) available for wing-net, one migrating eels in the river and 73- Number of fishermen : 6 location for coghill,. 56% in the lake. Fishing gear used : wing-nets Effort in the lake : 4 weirs or coghill in rivers, eel weirs in CPUE : kg/net night the lake. Regulation : open from late August to first week in January. River Shannon The fishery is located in the Yield : 10-15 T Two mark-recapture experiment main rivers and downstream gives a mean escapement of 60% of McCarthy and Effort : 20 river fykes 40 the lakes of the whole the migrating eels. Cullen (2000) nets in two locations for catchment. coghills Number of fishermen : 60 No measurement of effort Fishing gear used : river fyke for silver eel or coghill. Regulation : open from October to March.

13 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

3.5 Conclusion on impact of exploitation on spawner escapement

There is no single comprehensive study on the impact of glass eel exploitation on the net spawner production in any river system; there are just two studies on the impact of yellow eel fisheries; for silver eel, several case studies have been documented, but their relevance for the stock-wide escapement is unclear. In contrast, there are many well- documented studies of the impact of exploitation on the local stock, providing a good starting point for further analysis of the impact on spawner escapement.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 14

4 DENSITY DEPENDENT PROCESSES IN THE CONTINENTAL LIFE STAGES

4.1 Introduction

The understanding of density dependent processes is fundamental to fisheries management, since it determines the resilience of the stock towards exploitation. Upstream migration, sex ratio, growth and mortality are affected by the density of a local eel stock. Fishery pressure, recruitment, re-stocking and obstacles to upstream migration all impact density. Where, when, and the type of density dependence an eel is subjected varies with life stage. Additionally, the phasing of the exploitation in relation to the density dependent processes determines the way management targets can be set. Where exploitation precedes a density dependent stage, the density dependence will compensate for the loss due to exploitation. Density dependence is assumed to be a key factor in the glass eel fisheries.

4.2 Migration in relation to stock density

Eels are catadromous, highly migratory species, with marine spawning grounds well separated from inshore and inland growth areas. Invasion mechanisms of river systems are poorly documented and include at least four distinct behaviour patterns (Feunteun 2001): “founders” that colonise rivers until they settle in the first available suitable habitat they encounter; “pioneers” that migrate upstream to the upper boundaries of the system; “home range dwellers”, that establish in a given area for several months to several years; and “nomads” that are erratic eels performing a general upstream shift as they search for suitable areas to forage and or to settle. These behaviours are not mutually exclusive, any eel shifting from one behaviour to another, depending upon ontogenetic attributes (age, experience, morphology, physiological stage, etc.), population parameters which determine density dependant movements; and environmental parameters (food availability, temperature, flow, carrying capacity of the ecosystem, etc.).

Evidence of density-dependant migration has been reported in a small number of recent studies. The assumption is that when the carrying capacity of a habitat, a river stretch or a lake is reached, eels tend to move from (or through) saturated areas towards unsaturated ones. The spatial organisation and the structure of the eel population were tested in the Vilaine Catchment (Briand et al. 2000). Upstream migration in this river system was blocked in the estuary by a dam built in 1970 for pleasure navigation and drinkable water adduction. An eel ladder was built in 1995 and recruitment surveys have been carried out since then. Electrofishing surveys conducted in 1981 and from 1998 to 2000 enabled analysis of the effects of recruitment increase on density and spatial organisation of the population. In 1981, 11 years after the construction of the estuarine dam, the population appeared depleted. Average density was < 0.2 eels m-2. Older eels dominated the population, >2 years of age, while younger eels were very scarce. In 1998, 3 years after the construction of the eel ladder on the dam, the density had increased to 1.2 eels m-2 in downstream reaches (Fig. 4.1). In 1998, the age structure was much younger than in 1981 with a large dominance of age 1 groups in downstream reaches, but variable among habitats. In upstream reaches, at 50-100 km from the estuary, the age structure was less dominated by age 1 groups, but the proportion of age 1 eels tended to increase between 1998 and 2000. Contrarily, in downstream areas, the proportion of age 1 tended to decrease during this period, together with the density of the population.

This suggests a density dependent migration behaviour: age 1 groups being forced into the periphery of the high-density area (about 0.8 eels m-2). In 1999, the high-density area extended upstream by about 30 km. Stations dominated by age 1 groups then being located at the periphery of this extended high density area. The same extension of the density dependant area also occurred in 2000. In a number of stations of the high-density area, the population tended to become older and the density lower. This would suggest a “wave” type migration, which is in contrast to that reported by Ibbotson et al. (2002) for eels colonizing the River Severn where upstream migration was mainly through diffusion.

15 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Figure 4.2.1 Spatial structure of the eel population in the Vilaine catchment (France), in 1981, 1998 en 1999.

Shaded area: carrying capacity attained and density dependent migration takes place. Adapted from Briand et al. 2000.

Fig ure 4: Variations of the spatial structuration of the eel population in the Vilaine catchment between 1981 and 1998 and 1999. Dashed area: carrying capacity attained and density dependent migrations take place. Adapted from Briand et al. 2000. 1981

1998

1999

5 10 50 100 200 20 km Density nb . 100m-2 N Age 0 Age 1

> Age2 All ages

The existence of density dependent migration behaviour was also tested in a small river system according to the methodology developed by Pollard et al. (1987). Annual density variations were tested for 6 years in a set of 22 stations (Robinet et al. unpublished data). The relation between ln(d) and ln(d+1/d) where tested for the whole population and according to size class (Fig. 4.2). This correlation was used in a number of studies on density dependant movements in order to test time-series (Pollard et al. 1987). There was a significant correlation for all size classes and the relation suggests that, in this river, density-dependent migration behaviour occurred at densities of 25 to 30 eels m-2. However, density dependence appeared to be significant within size classes but independent from other size classes (Table 4.1).

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 16

Table 4.1: Density dependence relations in European eel population according to four size classes: 150 <150mm; 151mm< 300 <300mm; 301mm< 450 <450mm; 451mm< 450+. Multiple linear regression ∆ per each size class vs. Ln (Nt) of four size classes. The table indicates the regression coefficient value (R) and the p statistics (**p<0.01; ***p<0.001). R²; determination coefficient. ∆; interannual coefficient variation = ln (Nt+1/Nt) (according to Pollard et al. 1987). Nt, Total density per station (ind m-2) of all stages at year t (see text). Nt+1, total density per station at year t+1. R= regression coefficient. R², determination coefficient.

Ln (Nt 150) Ln (Nt 300) Ln (Nt 450) Ln (Nt 450+) R² ∆ 150 -0.603*** 0.139 -0.254 0.290 0.305 ∆ 300 0.152 -0.701*** -0121 0.115 0.408 ∆ 450 0.021 0.022 -0.600*** 0.060 0.274 ∆ 450+ 0.237** -0.194 0.186 -0.909*** 0.488

The existence of such a density dependant threshold could be used to determine whether the carrying capacity of catchments is reached or not. This could also be used to define restoration targets in various habitats.

7 Saturated

6

5 ]

Nt 4 [ 3 Ln 2

1 Unsaturated 0 -3 -2 -1 0 1 2 3 4 °

Figure 4.2. Density dependence of abundance variation in an European eel population of a small river, the Frémur. Nt: Total density per station (eels.m-2) of all size classes at year t (see text); Nt+1: total density per station at year t+1; ∆°: interannual coefficient variation = ln (Nt+1/Nt); R: regression coefficient; R²: determination coefficient. Adapted from Robinet et al. unpublished data.

4.3 Sex ratio and stock density

Considerable differences have been observed in the sex ratios of local populations, in space as well as in time. The general pattern is described as males dominating the dense populations in lower stretches of rivers, while females dominate the sparsely populated upper stretches.

Temporal changes in sex ratio within a river system have several long lasting implications that extend beyond ecological considerations and include several key fishery concerns; the production of the spawning population (primarily females) and the monetary value of the catch. Extreme changes in sex ratio from the natural state in either direction will affect one or both of these concerns. In the absence of significant density manipulation forces, sex ratios appear to be stable over time. In the absence of an eel fishery or apparent changes in recruitment in the Annaqutucket River, USA, sex ratios of American eels, Anguilla rostrata, have been consistent over a twenty-year period (Krueger and Oliveira 1999).

17 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Intense yellow eel fisheries or substantial changes in recruitment have been implicated in altering sex ratios from natural (historic) levels. Fisheries that target larger yellow-phase female eels may shift the silver eel sex ratio in favour of male eels. This is illustrated by the intense size-selective yellow eel fishery that began in Lake Ellesmere, New Zealand, in the 1960’s that has caused sex ratios of the shortfin eel, Anguilla australis, to shift from 1: 4.6 male to female to 235: 1 male to female by 1994 (Jellyman and Todd 1998).

Changes in recruitment have also been found to influence sex ratios. Evidence from an American eel stocking experiment suggest that changes in density of 0.5 m-2 can shift a population from predominately female to predominately male (Vladykov & Liew 1982). There is evidence that artificial enhancement (re-stocking) of Lough Neagh, Northern Ireland, has increased the proportion of male silver eels (Frost 1950; Parsons et al. 1977). However, the current stocking/sex ratio relationship in Lough Neagh does not appear to exhibit a density dependent relationship (Rosell, working group paper 2000, 2001). This situation will require another 5-6 years before the changes in stocking/recruitment over the last 10 years are realised in silver eel production. Decreased recruitment and/or increased productivity have also been associated with an increase in the proportion of female eels in the Burrishoole River system, Ireland, from 37.5 to 94.5% (Poole et al. 1990). A reduction in elver recruitment from 0.11 to 0.02 m-2 resulted in an increase in the proportion of female silver eels from 3:1 to 9:1 in the Comacchio Lagoons of Northern Italy (Rossi et al 1988). The increase in the proportion of females in the Baltic was also attributed to the decrease in elver recruitment (Svärdson 1976).

4.4 Mortality

For wild populations, the only published information is for the relatively low density Imsa River, Norway, where density dependent mortality was observed (Vøllestad & Jonsson 1986). Density-dependent effects on mortality have also been explored in stillwaters where the fishery is supported by artificial restocking of juvenile eel.

Stocking and yield data from a number of German stillwaters suggests a negative exponential relationship between mortality and density, see Figure 4.4 as an example. The analysis has assumed that one pigmented eel (bootlace) represents 5 glass eel and there is a lag of 7 or 8 years between stocking and recruitment into the fishery.

35.0

30.0

25.0 y = -6.0822Ln(x) - 14.138 20.0 R2 = 0.7319 15.0

10.0

5.0 Percentage recapture in year n+7 0.0 0 0.02 0.04 0.06 0.08 Re-stocking rate (glass eel equivalent per m2)

Figure 4.4. Percentage of eel recaptured in the Rangsdorfer See fishery against density of glass eel stocked seven years before.

4.5 Growth

Belpaire et al., (1989) and Klein Breteler et al., (1990) have shown a decline in growth rate with increasing density for eel stocked into ponds. In contrast, no significant difference in size among ponds was found for glass eel stocked into eight ponds at densities ranging from 0.016 to 0.16 glass eel m-2 (Klein Breteler 1992).

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 18

For riverine populations, Aprahamian (2000) found no relationship between growth rate and density or biomass. The study was carried out at 15 sites on seven tributaries of the lower Severn where the annual growth rate varied between 16.4 – 27.9 mm yr-1, the density from 0.12-1.14 eels m-2 and the biomass from 2.56–25.24 gm-2. There were significant differences in growth rate between sites (P<0.05), but these differences were not related to density or biomass (Figure 4.5 a&b).

a) )

-1 40.0

30.0

20.0

10.0

0.0 Growth rate (mmyr 0 0.2 0.4 0.6 0.8 1 1.2 1.4 Eel density (no. m-2)

b) )

-1 40.0

30.0

20.0

10.0

0.0 Growth rate (mmyr 0.0 5.0 10.0 15.0 20.0 25.0 30.0 Eel biomass (gm-2)

Figure 4.5 a&b The relationship between growth rate and density (a) and biomass (b) from 15 sites in the lower River Severn, UK, Aprahamian (2000).

4.6 Conclusion on density dependent processes

The presence of density-dependent effects is apparent in the life history of the eel and must be monitored/regulated at all life history stages. The association between density and the life history variables discussed offer two management options. The first is to manage the fishery in order to maintain a stable population density that ensures 30% SPR is achieved. The second option is to maintain historic population densities within each fishery. Both options will involve enhancing recruitment possibly by limiting glass eel harvesting by increasing access to suitable habitat, by restocking young eels thus altering the density levels at the early life history stages. In addition management of the yellow eel fishery will also be required.

At present it is not possible to quantify the effect of density on eel upstream migration, growth and mortality, for wild populations, for lack of information and some conflicting results. However, for those stillwater populations that are supported by artificial stocking of juvenile eel density dependent effects on mortality are evident and available information suggests common thresholds in large areas.

19 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

5 HABITAT LOSS

5.1 Introduction

Fisheries for eel are wide spread and the assessment of the impact of exploitation has a long tradition. Clearly, management of eel fisheries is an essential part of a management plan. However, sustainable management and a stock recovery plan should also take into account anthropogenic impacts upon the stock, other than exploitation. The decline of the eel in Europe is often related to the decline of its continental habitat, its accessibility and its quality. The relative magnitude of these factors, in relation to the impact of exploitation, has not been quantified, but it seems likely to be significant in many European countries. In this chapter, the loss of habitat will be assessed.

5.2 Dams

During the 20-century, large dams emerged as one of the most significant and visible tools for the management of water resources. More than 45 000 large dams and far more smaller dams around the world have played an important role in developing local communities and economies by increasing agricultural production, energy generation, flood control and domestic use. The member states of the European Union regulate the flow of 60–65% of the rivers in their territories, though the amount varies from basin to basin (www.dams.org). The number of new large dams built per decade sharply increased after the World War II (Figure 5.2).

Figure 5.2. Number of large dams commissioned by decade in Europe. Data from ICOLD World Register of Dams (http://www.dams.org). (*) Data for after 1990 is underreported.

Dams fragment and transform aquatic and terrestrial ecosystems with a range of effects that vary in duration, scale and degree of reversibility. These ecosystem transformations occur in the upper, lower and mid-reaches of watersheds, and they also impact on river estuaries.

The ecosystem impacts of dams can be classified according to whether they are:

⇒ first-order impacts that involve the physical, chemical, and geomorphological consequences of blocking a river and altering the natural distribution and timing of streamflow; ⇒ second-order impacts that involve changes in primary biological productivity of ecosystems, including effects on riverine and riparian plant-life and on down-stream habitat such as wetlands; or

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 20

⇒ third-order impacts that involve alterations to fauna (such as fish) caused by a first-order effect (such as blocking migration) or a second-order effect (such as decrease in the availability of plankton).

These problems may be magnified as more large dams are added to a river system, resulting in an increased and cumulative loss of natural resources, habitat quality, environmental sustainability and ecosystem integrity. The magnitude of river fragmentation can be very high. In Sweden, for example, only three major rivers longer than 150 km and six minor rivers have not been affected by dams.

As a physical barrier, the dam disrupts the movement of species leading to changes in upstream and downstream species composition and even species loss. River-dwelling species include anadromous fish such as salmon and catadromous fish such as eels. The World Commission on Dams (WCD) Cross-Check Survey found that impeding the passage of migratory fish species was the most significant ecosystem impact, recorded at over 60% of the projects for which responses on environmental issues were given. In 36% of these cases, the eventual observed impact of the large dam on migratory fish was not anticipated during project planning. Migratory fish require different environments for the main phases of their life cycle: reproduction, production of juveniles, growth, and sexual maturation. This partly applies also to the eels’life cycle.

Fish passes are often used as an engineered mitigation measure for reducing impacts on fish, but many studies show that fish passes are not available, not effective or not working at all. Even when fish passes have been installed successfully, migrations can be delayed by the absence of navigational cues, such as strong currents. This causes stress on the energy reserves of the fish such as eels or on the motivational cues to migrate and successfully colonise or transfer to upstream or downstream habitat. It enlarges the effects of density dependent processes in the local eel population downstream the dams or diminishes the contribution of the local population to reproduction of the stock.

5.3 Hydropower stations and pumping stations

5.3.1 Damage and mortality at turbines and pumping stations

Dams often are used for electricity generation by hydropower stations. Dikes are provided with pumping stations for clearing excess water. Hydropower stations and pumping stations cause damage and mortality to eels, specifically silver eels at their downstream migration. If the silver eels meet a hydropower turbine or a pump on their downstream migration, they can suffer damages or can die due to

• jam at the protection screen • collision with parts of the turbine • quick changes of the hydrostatic pressure • predators in the tailwater.

The rate of damage (%) depends on the position of the turbine in the river bed (eels migrate in the main current), the working regime (switching off the turbine during the main migration period reduces the damages), the efficacy of the protection screen, the turbine type, the water flow rate, the turbine type and on characteristics of the turbine.

5.3.2 Damage and mortality by passage through the turbine

Fishes passing turbines are susceptible to physical and physiological damages such as skin- or fin damages (24%), pop- eyes (< 1%), internal haemorhagues (12%), ruptured swimbladders and inner organs (< 1%), spinal fractures (12%) or simply (partially) being cut into pieces or decapitated (10%) (Holzner 1999).

For Kaplan turbines, the damage rate has been described by Von Raben (1957) and the relation has been modified by Berg (1985).

Francis turbines are much more dangerous for eels because of the smaller gape width in-between the blades. They are used mainly in smaller streams. The damage rates are roughly 2.5 times higher than at Kaplan turbines (Holzner 1999). At maximum water discharge, the mortality at Kaplan turbines is reduced and at Francis turbines increased (McCleave 2001).

21 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Damage rates in both types of turbines are linearly dependent on the length of the fishes. This follows the theoretical formulas from Von Raben (1957) and Berg (1985) but has also been shown in practise by Holzner (1999). The risk for female silver eels (size ranging from 50-100 cm) for damage by turbines encountered, is therefore higher than for male silver eels (sizes up to 50 cm) and much more higher than for salmonid smolts.

Table 5.3.2 gives a literature overview on mortality estimates of eels at hydroturbines.

Larinier & Travade (1999) reviewed eel mortalities in turbines and conclude that, due to their body lengths, mortality in low-head turbines is 15-30% (minimum 10% in large ones) and more than 50% in the smaller turbines installed in most of the small hydropower stations.

5.3.3 Damage and mortality by passage through a pump

Marshlands and riverine polders are frequently drained by pumping stations. The working principle of pumps is similar to turbines and a similar effect on eels is assumed. Germonpre et al. (1994) investigated the damage rate and mortality of screw pumps and centrifugal pumps. These were quite low:

Screw pumps: 13,2 – 25 % damage rate; 3,5 % mortality Centrifugal pumps: 1,4 – 3 % damage rate; no mortality.

5.3.4 Damages in the tailwater

An additional mortality factor is the loss by predators (birds, mammals and fishes) in the tailwater. The eels are disoriented after the turbine or pump passage and can be easily caught by predators. This loss at turbines may amount up to 33% (Späth 1998).

5.3.5 Damages before the turbine and protection devices

Protection measures, i.e. deflection screens, in front of the turbines may prevent eels being entrained into the turbines or pumps. Generally, gape widths ≤ 15 mm are necessary to stop silver eels entering a turbine or pump. An inclined screen ending with a bypass at the downstream end of the turbine is necessary to protect the silver eel from jamming against the screen. A screen inclination angle against the current direction of 15 ° is sufficient for the deflection and protection of the eels (Adam 2000, Adam et al. 2002). The mortality rate due to jamming against an insufficiently declined screen is 7 - 69 % and on the average mean 34 % (Rathke 1993, 1994, 1997 and Holzner 1999).

Deflecting screens have to be installed by laws in several European countries. Some countries do not have any legislation for the protection of eels. However, the protection measures implemented according to the actual laws are mostly insufficient (too large gape widths, no screen inclination or bypass demanded).

5.3.6 Cumulative mortality

Many rivers have more than one hydropower station. Downstream migrating silver eels run risks on mortality and damage each time they encounter a hydropower station. PRIGNON et al. (1998) made an estimation of the expected silver eel mortality due to the 6 consecutive hydropower stations on the river Meuse using the model of LARINIER & DARTIGUELONGUE (1989). The model predicted an immediate mortality varying between 34 and 45% for males and 40 to 63% for female individuals. Prognoses made for new plants in Flanders indicate an immediate mortality of ±17% per site. The cumulative effect of a hypothetical situation with 5 consecutive hydroelectric power plants would result in a mortality rate of over 60% without taking into account the postponed mortality. A study on 12 subsequent power stations at the upper Rhine between Schaffhausen and Basel gave an average total mortality of 92.7 % (DÖNNI et al. 2001). McCleave (2001) made a simulation of the cumulative effect of 13 power stations in the Kennebec River Basin (Maine, USA). The total water area amounts to 27,000 ha. According to his model and without any fishing mortality or the effect of the water regulation weirs, the following proportions of female eels reach the open sea: at 10 % mortality per power station: 63 % at 20 % mortality per power station: 40 % at 40 % mortality per power station: 18 %.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 22

5.3.7 Hydropower generation and capacity: status, prospects and impacts on the European eel stock

The total electricity generation and the relative contribution of hydropower widely differs between the European IEA countries (Table 5.3.7). Compared to other continents, the share of hydropower to the total electric energy (31%) is relatively large (Fig. 5.3.7.1). Most of the hydroelectric energy that is produced in the countries that are important for the eel production occurs in Norway, France, Germany, Finland, Italy, Portugal, Spain, Sweden and Turkey. In all of these countries but Norway hydropower produces a serious but unquantified loss of silver eels of the local stock. In Norway most of the local stock is produced in the marine habitat that is not affected by hydropower. The IEA prospects are that only Turkey will grow in hydroelectric capacity in the near future. But a recent European Directive (2001/77/EC) is likely to enhance the construction of new hydroelectric power stations over Europe. European Community member states have to increase the proportion of electricity produced from renewable energy sources from 13.9% to 22% of the gross national electricity consumption. This possibly leads to a considerable extension of the hydropower usage, even in very small rivers. In the UK more than 991 potential sites have been identified for small- scale hydropower production (www.small-hydro.com). In Belgium, over 150 new places are envisaged. Most of these stations will have small capacity. Small capacity power stations usually have a small energetic gain but may induce considerable ecological damage, while the costs of compensating measures is not related to the capacity.

The number of existing small hydropower stations seems to be large. Germany e.g. has more than 6000 hydropower stations < 5-10 MW, with a total capacity of 1.3 GW. 4881 of these are less than 1 MW and generated 1.49 TWh in 1996, representing only 9.2% of the total hydropower production. There is a potential for another 1000 small hydropower stations (Bunge et al. 2001). It appears from that study that small hydropower stations (< 0.1 MW) are not able to produce energy at current market prices. Probably the same applies to other countries such as France (1700 small hydropower stations) and Italy (2700 stations).

Distribution of existing large dams by region and purpose. Source: Adapted from ICOLD, 1998

Figure 5.3.7.1. Share of hydropower to total electric energy in different continents.

5.4 Habitat loss

Irrigation is the single largest consumptive use of fresh water in the world today. It is linked to food production and food security. About one fifth of the world’s agricultural land is irrigated, and irrigated agriculture accounts for about 40% of the world’s agricultural production. The total area irrigated expanded dramatically during the first years of the green revolution in the 1960s, increasing yields and bringing down food prices. From 1970 to 1982, global growth in the irrigated area slowed to 2% a year. In the post green revolution period between 1982 and 1994 it declined to an annual average of 1.3%. An estimated 30 to 40% of the 268 million hectares of irrigated lands worldwide rely on dams. Discounting conjunctive use of ground water and surface water, dams are estimated to contribute to at most 12-16% of world food production.

On the Garonne – Dordogne- Charente basins in France e.g. there has been a 5-fold increase of irrigated areas, mainly for maize culture, since 1970 to 2000 (roughly from 100,000 hectares in 1970 to 500,000 in 2000) (Teyssier et al. 2002). For the Adour basin there has been a 4-fold increase of the irrigated maize culture from 1980 to 1998 (Baudry 1999). At times the riverbed of the Adour river is completely dry by withdrawal of water for irrigation purposes (Prouzet, pers. comm.). This poses a problem for survival of eels.

23 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

More or less the same applies to the large polder areas in the Netherlands, where water is withdrawn during wintertime as a prevention for floods due to precipitation. The eels run higher risks then for drying out or for freezing.

It is estimated that more than half of the total area of wetland disappeared in France during these last thirty years. That regression is continuing. For example 18 of the 19 main wetlands of the drainage area of the Adour-Garonne basin had suffered degradations resulting in habitat loss for eels (De Faveri 2002).

5.5 Case studies

5.5.1 Villaine

The construction of the Arzal dam eel ladder in the Villaine estuary in France, together with fisheries management measures, allowed to enhance fluvial recruitment from near zero -limited to yellow eel having the ability to cross the dam overflow- to a level ranging between 0.2 to 2.4 millions of glass eels per year. Ladder installation resulted in an increase of density of eels by a factor 6, and an increase of density of age 0 and 1 eels with a factor 14 and 30 (Briand, pers.comm.). Figure 5.5.1.1 illustrates the high fragmentation of the habitat due to numerous dams in the Villaine drainage area.

Chevré

Cheze Oust Canut Vilaine Arches CanutNord Aron Arz Sud Trévelo

Arzal

10 km N

Figure 5.5.1.1: Map of the Vilaine watershed, 1998 and 1999 monitoring. ▲= dams, ● = electrofishing station

5.5.2 Garonne-Dordogne-Charente

Another example of the high fragmentation of eel habitat by dams is provided by the Garonne-Dordogne-Charente area (Figure 5.5.2.1).

Generally, the fishpasses installed on dams are adapted to migratory salmonids and shad but are not suitable for eel. On the Dordogne river an experiment was performed in 2000 by the GHAAPE/MIGADO to compare the efficacy of two types of fishpasses on the upstream migration of eels at the “Tuilières” dam : one adapted to the passage of large migratory fish and the other adapted to glass-eel behaviour (brush-carpets). The results showed clearly the inefficiency of the salmonid fish-passes for the passage of elvers on large dams. Many of these dams are combined with electricity generation by hydropower (Figure 5.5.2.2)

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 24

Figure 5.5.2.1: Densities of dams for the upstream migration on the Garonne-Dordogne-Charente basins (after Teyssier et al 2002)

25 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Figure 5.5.2.2: Densities of hydroelectric dams for the downstream migration on the Garonne-Dordogne-Charente basins (after Teyssier et al 2002)

5.5.3 The Frémur

The Frémur (Figure 5.5.3.1) is a small river of northern Brittany (France) which opens into the British Channel next to Saint Malo. Its catchment covers about 60 km² and the overall length of the river and its tributaries is 45 km, comprising 17 km for the main stream. The slope varies between 0.1% and 2% for an average of 0.6%. Despite its small size, the Frémur provides for a wide range of habitats from high velocity streams of the trout zone to lentic waters of the bream zone in downstream areas, man made ponds and reservoirs, wetlands, etc. The total extent of water is 75 ha including 5 ha of running waters (streams) and 70 ha of still waters (ponds and reservoirs). Therefore, this river appears to be representative of small coastal catchments of Western France.

A total of 6 dams and weirs are present on the catchment. Three of them were totally impassable until they were equipped with eel passes (Feunteun et al. 1998). The most important one, Bois Joli, was built in 1991. It is 14 m high and creates a 3 million m3 reservoir to provide the county with drinking water. Most of the time, it is not full and the flow is controlled by pipes. When it is full, the water flows over a discharge weir. It has been equipped with an eel lift to restore upstream migration. The Bois-Joli dam was also equipped to increase downstream migration.

A survey plan was started in 1995 in order to estimate the effects of equipments on recruitment, population characteristics and production of silver eels (i.e. Acou 1999; Feunteun et al. 1998; 2000; Acou et al. 2001; Guillouet et al. 2000; Legault et al., in press). To achieve this goal a research project was conducted and used complementary sampling and analysis techniques. Exhaustive daily follow-up of recruitment was conducted since 1995 using a trap derived from the eel lift (Guillouet et al. 2001). Population structure was analysed with an intensive electrofishing

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 26

survey combined with a spatial analysis and a set of mark recapture experiments (Feunteun et al. 1998; Guillouet et al. 2001). Downstream migration was surveyed on a daily basis since 1996 using a Wolfe trap enabling total catches of emigrating silver eels.

35000 1400

30000 1200

25000 1000

20000 800

15000 600

10000 400

5000 200

0 0 1996 1997 1998 1999

Recrues Stock Dévalaison

Figure 5.5.3.1. Yearly variations of recruits (pink) stock (blue) and downstream emigration (yellow)

The interpretation of this study were:

• Without the eel passes, the eel population would have progressively disappeared.

• The eel pass showed considerable variations of recruitment (mainly elvers), but did not provoke significant variations of stocks and of silver eel production.

• The eel passes thus enabled to sustain stocks at saturated levels and production of about 1000 silver eels per year, i.e. 13 silver eels per ha of habitat and per year.

The escapement of silver eels was only possible over the weir once the Bois Joli reservoir was full. Generally the emigration runs occurred later than in neighbouring open systems and was postponed each year by 1 to 6 months. Passage through the barrage was only possible for eels finding the entry of a small pipe that ensured minimal legal water discharge. This passage provoked total mortality. It was modified in order to reduce mortality (Legault et al. Silkeborg poster). It enabled passage of approximately 10% of the total runs (about 100 eels a year).The outcome is that downstream migration of silver eels, and mortality through barrages is easily reducible.

On the Atlantic coast of France, many salt marsh estuarine systems have been reclaimed for salt production since the 11th century (Feunteun 1994; Feunteun et al. 1999). This resulted in the existence of a dense network of ditches and water bodies which provide about 180 km2 of highly suitable habitats for eels. These reclaimed marshes extend over a wide territory from the Vilaine estuary to the Arcachon Bay. They host high densities of eels ranging from 30kg to over 300 kg per ha (average 60kg per ha) (Feunteun 1994; Feunteun et al. 1999; Baisez et al. 2001; Baisez 2001). Eels appeared as the most adapted species to exploit such a constraining habitat which are characterised by rapid variations of water levels, temperature, oxygen concentration, Salinity, etc. Therefore low interspecific interaction occurred in these areas, favouring eel growth, survival and more generally population dynamics.

27 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

In some areas, new habitats were created in reclaimed wetlands for fish and game conservation (Eybert et al. 1998). Newly created habitats were colonised by dense populations of eels within a few months. Such increase of habitat availability was thought to reduce mortality, improve growth, and finally eel population size through density dependant mechanisms and therefore was interpreted as an appropriate method to increase spawner production at the catchment level.

Therefore, it is recommend that such kind of network should be created in reclaimed wetlands, both in marine area, or in flood plains in order to provide new highly suitable habitats for eels. In eel saturated systems as coastal rivers of the Atlantic, saltmarshes, coastal lagoons, floodplain wetlands of southwestern Europe (France, Spain, Portugal) such a restoration/creation option appears to be a highly efficient option to increase production of silver eels by an aquatic ecosystem, at the catchment level. An other outcome of this is that wetland habitats ought to be protected, managed and maintained at least at the actual level in order to conserve highly suitable habitats for eels.

5.5.4 The Meuse

An important number of migration barriers are present in the river Meuse from the headwaters in France to the estuary in the Netherlands. Part of these weirs are equipped with fish ladders, turbines and/or fish deflectors (Table 5.5.4.1).

Table 5.5.4.1. Dams, hydropower stations and fishpasses in the river Meuse mainstream. Source : International Commission of Protection of the Meuse (www.cipm-icbm.be)

Number of barriers With fish ladder With turbine Turbines with fish deflection France 23 2 10 0 Belgium, Wallonia 15 11 6 1 Belgium, Flanders 0 0 0 0 The Netherlands 8 5 2 (+3 planned) 0 (+3) Total 46 18 18 (+3 planned) 1 (+3)

Due to those migration obstructions a large surface of potential yellow eel habitat is not accessible for eels, especially in the upper reaches of the catchment in Flanders, Wallonia and France. And survival of the local Meuse eel stock is negatively affected by the turbines.

The Benelux migration decree on the free migration of fish species in the hydrographic river basins of Benelux countries was signed in The Hague in 1996 and stated that all Benelux countries have to guarantee the free migration of fish species in all water courses. Priority must be given to the large catadromous and anadromous migrators to allow the migration between spawning areas and growth habitats. By January 1st, 2010 migration in all fish species in all waterways should be made possible irrespective of the manager (Benelux 1996). Also France endorsed the idea to make all waterways of the Meuse catchment suitable for free fish migration.

Consequently, a number of initiatives were taken to enhance possibilities of fish and eel migration over the Meuse catchment. Also initiatives for habitat restoration have been realised or are planned in the near future all over the Meuse catchment (Table 5.5.4.2). These measures are likely to have a positive impact on yellow eel habitat. However, plans for building additional new hydroelectric power stations on the silver eel escapement routes can reduce the positive effect of the restoration measures.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 28

Table 5.5.4.2. Restoration measures in the river Meuse. Source : International Commission of Protection of the Meuse (www.cipm-icbm.be)

Country Region Management or restoration period responsible measure France Haute Marne Restoration of 10 meanders 1992 Fisheries association Haute Marne Lorraine Fish ladders 1999-? Agence de l'eau Ardennes Restoration of dead arms - 1998-1999 Associations de pêche habitat diversification Ardennes Monthermé fish ladder 2000-2002 Voies Navigables de France Belgium Upper Meuse Conservation of river arms 1991-? MET, Contrat de Rivière Wallonia (managing connections) Upper Meuse Fish ladders Restoration of river MET continuity Belgium Small river Taking away migration bariers 2000-2010 Province of Limburg and water Flanders projects managers Bosbeek Restoring meanders 2000-2003 Exploitation committee The Limburg Borgharen, Linne, Roermond, Planned or Directie Limburg Netherlands Belfeld, Sambeek Realised Fish ladders Limburg Groeningen River arms Realised Directie Limburg management Limburg Sambeek veer River banks Realised Directie Limburg Nijmegen- Afferden-Ossenkamp river bank Dienstkring Nijmegen-Maas Maas restoration

Nijmegen- Gebrande kamp Grontmij Maas Nijmegen- Grave fish ladder Planned Ballast/ Dir. Limburg Maas Nijmegen- Lith Fish ladder Realised Dir. Limburg Maas Zuid-Holland Hoogezandsche Gorzen River planned Rijkwaterstaat banks Zuid-Holland Ventjagersgaatje River banks Realised Rijkwaterstaat Zuid-Holland Ezelsgors River banks Realised Rijkwaterstaat Zuid-Holland Haringvlietsluizen Sluices Planned Rijkwaterstaat management

5.5.5 The Rhine

The total number of hydropower plants in the river Rhine mainstream is 20, (10 in Germany-France and 10 in Switzerland). The Mosel, Lahn, Main and Neckar are important tributaries to the Rhine and formerly completely accessible for and used by the eel. The Mosel counts 22 hydropower stations. There are 8-9 hydropower stations situated in its downstream section up to the first two important tributaries Kyll and Sauer. The Main counts 34 hydropower plants, of which 6 are situated in the downstream section. In the Neckar there are 10 hydropower plants in the downstream section and >10 in the part upstream its first locally important tributary. And in the Lahn there are more than 10 hydropower stations (Muyres, in prep.). Indicative for the high fragmentation of the Rhine-tributaries (MUNLV 2001) are the maps of dams and hydropower stations in the rivers Sieg and Wupper (Figure 5.5.5.1).

The International River Commission of the Rhine produced an inventory of measures undertaken and underway for restoration and conservation of the Rhine (IKSR 2001). These are measures on the improvement of the ecosystem, flood prevention, waterquality and groundwater conservation. The measures with regard to improvement of the ecosystem are specified for the upmost, upper, middle and lower Rhine. Among these are lowering of summerdikes (> 20km2 per section), reactivation of dammed old riverbranches and connection of floodpains with the mainstream (> 25 riverbranches, additional dredging etc. included), restoration of riparian transects (> 400 km, gravel, vegetated and wooded floodplains included), construction of fishpasses at existing powerstations and dams (in the mainstream and in the tributaries making part of the diadromous-fish program), nature improvement of > 3500 km tributaries.

29 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Figure 5.5.5.1 Fragmentation and hydroturbines in the rivers Sieg (top) and Wupper (bottom), tributaries of the Rhine (after MUNLV 2001),

5.5.6 North America

The Atlantic States Marine Fisheries Commission (ASMFC) provides a thorough review of quantifying stream habitat in their Fishery Management Plan (FMP) for American eel. This review is presented below with minor modifications (ASMFC 2001).

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 30

Busch et al (1998) used an ecosystem health assessment approach, developed for the Lake Ontario watershed, to determine that Atlantic coastal streams from Maine to Florida have 15,115 dams that can hinder or prevent upstream and downstream fish movement. This results in a restriction or loss of access for fish to 84 percent of the stream habitat within this historic range. This is a potential reduction from 556,801 km to 90,755 km of stream habitat available for migratory and diadromous species such as American eel. The analyses were based upon the regional boundaries established by the USEPA database and excluded obstruction caused by most natural barriers.

By region, the potential habitat loss was greatest (91%) in the North Atlantic region (Maine to Connecticut) where stream access is estimated to have been reduced from 111,482 km to 10,349 unobstructed km of stream length. Stream habitat in the Mid Atlantic region (New York through Virginia) is estimated to have been reduced from 199,312 km to 24,534 km of unobstructed stream length (88% loss). The stream habitat in the South Atlantic region (North Carolina to Florida) is estimated to have decreased from 246,007 km to 55,872 km of unobstructed stream access, a 77% loss.

In the U.S. portion of the St Lawrence river, 455 dams result in 24,693 km of stream habitat lost or restricted from a total of 30,085 km (82% loss) to migratory fish originating in or having Lake Ontario as their destination. Since dams on the St. Lawrence River hinder fish movement through the St. Lawrence River to and from the Atlantic Ocean, the total km of stream access lost or restricted in the Lake Ontario and St. Lawrence River watershed is actually much larger.

The dam database used by Busch et al. (1998) included information on dam heights. It identified 3,512 dams in the North Atlantic Region of which 448 are less than 10 ft. high, 2,260 are between 10 and 24 ft. high, and 813 are higher than 25 ft. Of all the dams, 561 are used for hydropower production. The Mid-Atlantic Region has 4,650 dams of which 475 are less than 10 ft. high, 2,563 are between 10 and 24 ft. high, and 1,603 are higher than 25 ft and 217 dams are used for hydropower production. In the South Atlantic Region, the 6,944 dams identified included 194 that are less than 10 ft. high, 3,993 between 10 and 24 ft., and 2,818 higher than 25 ft. Of the dams in this region, 141 are used for hydropower production. Dams in the US Lake Ontario basin include 64 that are less than 10 ft. high, 238 that are 10-24 ft. high, and 153 that are 25 ft. or higher. Hydropower production was the use identified for 181 dams.

Dams that require special licenses such as for hydropower production or navigation provide opportunities for fish passage if required by the resource management agencies. However, only 1,100 were identified for hydropower production and 50 for navigation out of the total number of 15,570 identified dams. Therefore, only 7% of these dams are covered by regulatory programs that could provide fish passage.

5.6 Conclusions on habitat loss

• Dams, weirs and dikes limit or reduce the upstream migration of eels on a large scale in Europe, but for instance also in North America. This enlarges density dependent mechanisms in the downstream areas. Most of the large dams have been built since the 2nd world war.

• Hydropower stations add a considerable mortality factor for downstream migrating silver eels. Mortality due to hydropower can occur at non-adequate deflection screens, in the turbine and in the tailwater.

• At the current state of hydropower technology, the elimination of the fishery or other mortality factors will not be sufficient to ensure the minimal SPR of 30 % in water courses with more than 3 or 4 subsequent hydropower stations.

• Although a quantification of the effect of hydropower on the European eel stock is lacking, the available information indicates a serious impact on mortality of the spawner population.

• Large-scale reductions of wetlands and diking has resulted in severe habitat loss for eels, both in preferred downstream areas and also in upstream riverine areas. Many restoration measures are currently taken in some rivers, but this only has importance for these local stocks.

• Protection measures for migrating eels at hydropower stations are essential, such as adequate deflection screens in front of the turbines and switching off the turbines during the main migration period.

• Working eel passes are needed at dams reducing dispersion of eels.

• More habitat loss for eels should be evitated.

31 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

6 REFERENCE CONDITIONS, EXPLOITATION AND HABITAT AVAILABILITY.

6.1 Introduction

Over the past decade, there has been increasing concern over the state of European eel stocks. The first widely noticed problem was a consistent and pan-European sharp decline in numbers of glass eel recorded at index monitoring stations in the early 1980s. The stock is considered to be outside safe biological limits. Long-term and consistent declines are now well documented in all phases of the life cycle over the entire distribution area, with most local stocks at historically low levels. Current scientific knowledge is inadequate to derive management targets specific for eel. Therefore, it has been advised (ICES 2001) to apply the universal reasonable provisional reference target of an escapement of spawners from the fisheries of at least 30 %, relative to unexploited conditions. However, this advice considers exploitation only, while decline in spawner production due to loss of habitat is probably as influential in many areas (see chapters 5 and 7).

The diversity in fishery exploitation practice, related to eel population structure and differing relative abundance of glass, yellow and silver eel makes a co-ordinated management response to the problem extremely difficult in the short- term. In particular, the tendency for those southern countries where there is traditionally high glass eel immigration, to exploit these recruits directly for consumption and, more recently, sale to aquaculture, creates potential conflict with those northern countries with traditionally low glass eel immigration and where eel are exploited at the grown yellow or emigrating silver stage. In addition to the differences in exploitation pattern, the growing phases of eel in these two extremes differ, regulated largely by the effect of temperature variation on growth.

The density of recruitment in areas traditionally focused on glass eel exploitation is much higher than elsewhere. It is assumed recruitment far exceeds carrying capacity of the inland habitats (see chapter 4). Consequently, unlike in northern areas, the amount of habitat available for the incoming recruitment is one of the key elements in local stock dynamics. Consequently, management of habitat and restoration of lost areas will be prime elements in local management schemes.

In the yellow to silver eel phase, the northern extreme is characterised by long freshwater lifespan (up to 30 years and more), low natural mortality, low growth rates and generally larger size of spawners, while at the southern end short freshwater lifespan (down to 5 years or less), high growth rates, high mortality rates and smaller size of (particularly female) spawners occur.

Against this background, the search for internationally agreed and scientifically sound methods of regulating exploitation and/or otherwise is obviously difficult. While there must be a common objective of protecting the whole panmictic stock, management measures aimed at implementing this objective will differ between the northern type recruitment and exploitation pattern and the southern one.

6.2 Reference conditions for northern type stocks.

In low recruiting stocks where exploitation is at yellow or silver eel stage, it makes sense to consider managing this exploitation by controlling Fishing Mortality and other anthropogenic mortality between recruited glass eel settled into growing areas and eventual emigration as silver eel. This requires some knowledge of the mortality profiles of fished stocks in comparison with unfished stocks, leading to life-table measures of F and M as a basis for proposed regulation.

The best possibility for mortality based descriptors of theoretical unexploited structure in stocks of growing eels may lie in the use of cumulative mortality from glass eel to emigrating spawner, rather than in attempting direct comparison between instantaneous annual mortality in different areas. One suggestion is the possible derivation length based mortality measurements or estimates which could allow direct application of common models across Europe – but a specific study comparing life-table measurements across a wide range of countries would be needed to test the usefulness of this idea

The use of M and F+M based characterisation of local eel stock has been extensively discussed in the 2000 and 2001 Working Group reports, derived or extrapolated theoretically from a few examples of (Mainly American eel) life-tables. The few additional data available to the 2002 group do not merit extensive re-modelling or derivation of any new definition of unexploited reference conditions based on life-table mortality. On a positive note, several newly commissioned studies in Europe may yield useful data within the next 3-4 years. Due to the extent of work involved in working up new data to life-table based M estimates, this area of study might need to be taken up by WG members inter-sessionally rather than be left to WG meetings.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 32

Examples of unexploited eel populations are not considered to be typical of any potential unexploited state in those systems currently fished. This is due to the reasons for the unexploited state – those stocks are usually not suitable for economic exploitation due to reasons including inaccessibility, oligotrophy (leading to small stock size, low density and low growth rates), or severe under-recruitment causing low stock density.

There are, however, some documented examples of unexploited or very lightly exploited eel sub-stocks (e.g. Svedang 1999, Poole et al 1990, Poole and Reynolds 1996), that might give some indication of characteristics suitable for incorporation into management objectives for stock restoration. Matthews et al (2001) described some length-frequency distributions from the Erne system, Ireland both compared between adjacent fished and unfished lakes and before and after periods of intensive fishing. Similar examples for Brittany, France are available (Sauvaget 2001, Briand 2001) However, population structures of unfished systems over Europe as a whole tend not to share many common characteristics and, in terms of internal population structure, range from systems where there are clear gradients of increasing size and increasing proportion of females with distance inland, to systems where there are no clear spatial differences in size distribution.

Size structure of eel ( Fouesnant coastal marsh Brittany)

10 9 8 7 Nu 6 mb 5 er 4 3 2 1 0 400 440 480 520 560 600 640 680 720 760 800 840

Size (mm)

Size structure of the Natural reserve of the Golfe du Morbihan

140

120

l 100 e e 80 of er

b 60 m u

N 40

20

0 200 250 300 350 400 450 500 550 600 650 700 750 800 850 900 950 Size (mm)

Figure 6.2 Comparison between fished and unfished Length-frequency (from Sauvaget 2001) Fyke net experimental fishing in coastal marshes (natural reserves)

In order to develop M-based targets derived from actual or theoretical unexploited scenarios, comparisons will need to be made with fished systems nearby in geographical terms and as similar as possible in terms of recruitment history and productivity. There is little potential for use of some notional standard fished system as the nature of exploitation patterns varies between countries and regions. For example, different minimum sizes and market preferences apply for yellow and silver eels. Some countries ban glass eel fishing or only permit it under special licence for restocking inland waters (e.g. Ireland). In contrast, there are areas (e.g. Baltic coasts) where glass eels are rarely observed and fishing is almost exclusively for silver eels.

33 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

In the application of an internationally agreed area-based spawner escapement target as proposed by the 2001 ICES/EIFAC WG, local conditions will have to be taken into account in the selection of areas over which the target should apply. One clear example of this is where hydropower turbines in lower reaches of rivers cause significant silver eel mortality. In such circumstances a country or region might wish to group catchments into a single management unit so that eels which would be subject to turbine mortality if allowed to escape are permitted to be totally exploited upstream of hydro stations, with compensating protection of silver eels from other catchments to ensure that a regional rather than individual catchment escapement target is met.

Low levels of current recruitment present a further problem to the definition of reference conditions for stocks. Given that almost all recruitment indices demonstrate that numbers of immigrating glass eels since the 1980s have been one tenth of numbers in previous decades, it is difficult and perhaps unwise to extrapolate from eel populations derived from current levels of recruitment to what might occur if recruitment recovered to former levels. This caveat also applies to the potential use of old population or stock data for setting reference conditions relevant to current recruitment. Given the recruitment history, it seems highly improbable that density dependent processes (see chapter 4) are occurring in to any significant extent in currently observed northern type eel populations, at any level of exploitation, other than in a few isolated examples, but may have had an influence for populations derived from pre 1980 recruitment. Management advice on what to look for in unexploited populations should nevertheless take account of the possibility of density- dependent processes (growth, mortality and migration) occurring should recruitment return to former levels

6.3 Southern type fisheries and stocks of growing eel

In those areas where recruitment still exceeds the capacity of the available habitat for potential spawner production, density dependent processes are likely to restrict survival of the glass eel to yellow eel stage (see chapter 4). Positive recruitment change could conceivably make no difference to spawner output without opening access to or restoration of habitat. In these areas, local managers should at least aim to meet a minimum target of using the total available potentially productive habitat. There are recent examples where this approach is being applied successfully, for instance in the Vilaine, France (Briand 2000), where progressive invasion of formerly understocked upstream habitats has followed fish pass construction and effort restrictions on the estuarine glass eel fishery. Where there is still excess glass eel following such measures, ideally, more ambitious targets should be applied, especially where habitats are degraded rather than simply inaccessible. In this case, habitat extension or recreation schemes have the potential for a positive benefit to spawner escapement for the whole stock. In other, (generally northern) areas, habitat extension will not necessarily have any benefit, as these areas, have never been known to have a saturated eel population (e.g. Scandinavia).

6.4 Possible levels of habitat restoration targets

Discussion of different levels of habitat restoration targets is at a relatively early stage, but it is clearly possible to envisage more ambitious targets than those above. A structured approach to habitat restoration targets could include the following levels:

1. Make full use of the entire existing habitat without modification. 2. Restore all the habitat where relatively little effort is required 3. Restore the habitat to be able to carry all the available recruitment 4. Restore habitat to some historical reference level 5. Restore habitat to some notional pristine condition.

In discussing these issues, it is likely that there will be convergent thinking between eel fishery management interests and consideration of implementation of the EU Water framework directive, particularly where there is national and international overlap between fisheries and environment/nature protection agencies.

6.5 Use of stock density measures in definition of reference conditions

Assessment of temporal and spatial stock density variation (e.g. by electrofishing or netting techniques) may yield an additional useful means of assessing population change (Knights et al 2001), taking into account local productivity and carrying capacity, but must be treated with care for a number of reasons. There are important potential interactions between state of exploitation, recruitment stock density and habitat quality or area available, which need to be taken into account in guiding managers on desirable population state. In particular, both exploited and unexploited sub-stocks may be constrained by the extent or quality of available habitat. Should habitat area or quality be restored without recruitment increases eels might become more dispersed and the stock could appear to be in a poorer condition, despite possible benefit to spawner escapement. In exploited systems targeting yellow eel, CPUE might fall in such scenarios, despite improvements in individual survival rates. Similarly, where recruitment is critically low, opening of new habitat,

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 34

for example by the use of fish passes, could also conceivably cause apparent reduction in local stock densities downstream of former obstructions. Intervals between surveys in temporal comparison of stocks must also be sufficiently low to allow detection of change due to recruitment change or new management practice – for example, recent data from the Vilaine, France (Briand et al. 2000) demonstrates inter-annual population expansion following Glass eel fishing effort restriction and eel pass construction.

Theoretically, the Water Framework directive will require Countries to collect density or stock biomass information on all freshwater fish stocks, including eel. However, the Eel WG is not hopeful that implementation of the WFD will be sufficiently uniform to allow the assumption that eel data suitable for national and international management of stocks will automatically result. Therefore, eel stock assessment for international reporting and management, while it need not be a necessarily completely separate exercise should not be allowed to depend on collection of data for WFD purposes.

6.6 Conclusion on reference conditions

For the eel population as a whole, a provisional management target has been advised of an escapement of spawners of at least 30 %, relative to pristine conditions. Implementation of this global objective in local management situations will require translation of the global objectives into local management targets. Several options are available, for which further development will probably be worthwhile. Additionally, a stepwise approach of the final objective might be recommendable, related to the complexity and involvement of the measures required.

35 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

7 EXPLOITATION VS. OTHER MEASURES

7.1 Introduction

The European catch of yellow and silver eels in marine and freshwater is estimated in the order of magnitude of 20 000 ton/year. Fishery regulations can allocate part or the whole of this as an increased escapement, to boost the spawning biomass and strengthen recruitment. With eels there are several other anthropogenic or otherwise controllable factors that can be used to reach an increased escapement. The most important measures besides regulation of the fishery are changes of:

1. Turbine mortality 2. Habitat accessibility 3. Predation 4. Restocking

Data availability to quantify the effect of changes in those factors is limited. This analysis tries to make first-order estimates. For 1 and 2 the primary data will be freshwater areas potentially or actually available to eels. Moriarty and Dekker (1997) made a compilation of this for parts of the distribution range (Sweden, Denmark, Germany, N Ireland, Rep Ireland, GB, Netherlands, France, Portugal, Spain, and Italy). Those countries represent a total eel catch of 14 500 ton/y (FAO statistics), which will be the baseline for comparison.

The current analysis is based on currently available information. The calculation on turbine mortality and habitat use preliminary data on the extent of waters affected, for which chapter 5 presents a thorough but not yet complete analysis.

7.2 Turbine mortality

The dataset in Moriarty and Dekker (1997) gives the total area of freshwater obstructed by dams to approx. 2000 km2. We assume that most of those obstructions are hydropower dams. In addition there are large areas that were counted as accessible in this survey because of eel ladders allowing upstream passage of eels. Sweden has the largest still freshwater areas in Europe (65 % of the total accessible areas or 12 000 km2) and it is known that essentially all of those areas are above hydrodams. A conservative estimate of the total freshwater areas where eel are present above hydrodams is 15 000 km2. An average European potential yield of 100 kg/km2 in freshwater (Moriarty and Dekker 1997) can be used as estimate for the production of silver eels. The mortality rate during turbine passages varies with turbine type and other factors. Typical rates are in the range 20-40%. Another unknown is the number of turbines that has to be passed on the way to the sea. We assume a range of 1-2 as a low estimate.

The resulting overall mortality with those assumptions is from 2500 ton/y (1 dam 20% loss) to 10 000 ton/y (2 dams 40%). This means that turbine losses can be about half of the total eel fishery. Turbine losses are essentially only of silver eels, and the probability of damage increases with length which makes the mortality selective to females. In this way the turbine loss is relatively more detrimental to the population recruitment than the fishery, which to a larger extent targets yellow eels and males.

The assumption of 100 kg/km2 silver eel production presupposes that the whole area actually is recruited with eels. This is probably not the case, which means that the above estimates in this respect are upper limits. Another uncertainty factor is to what extent downward eel passages exist at the hydroelectric dams. The trend has been to remove fine meshed grids from water intakes so the proportion of dams with functioning silver eel passages is likely to be low.

7.3 Habitat accessibility

With the provision that downstream passage is guaranteed the installation of eel ladders that allow small eels to colonise freshwater areas above dams is an action that will increase escapement. The area compilation in Moriarty and Dekker (1997) shows that 1 200 km2 of still water and 700 km2 of rivers presently are blocked from eel ascent. Using the same yield estimate as above we find that eel ladders could increase the escapement by 2000 ton/y if those areas were kept unfished.

In addition to this large areas of lakes and wetland have been lost over history by draining and water level reduction of lakes. No European-wide quantitative estimates are available for this but case studies are discussed in chapter 5.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 36

7.4 Predation

Predation mortality on eels due to other fishes and cannibalism is part of the management of the fisheries. Other predators exist however – birds, marine mammals – and management of such species will influence the population development of the eel.

Bird predation has been studied for cormorant (Phalacrocorax carbo) in particular. Herons (Ardea cinerea) and other bird species eat eels too. The effect of a Heron colony on an eel population was estimated in the west of France (Feunteun & Marion 1994, Marion et al 1998). This study showed that herons were opportunistic, and therefore their diet composition was similar to that of the available preyed fish community. A more or less constant predation rate of 11% was noted, in other words herons in the feeding territories predated 11% of the standing stock of eels (i.e. 6kg/ha of eels consumed per ha and per year). This predation rate was locally higher outside territories, where concentration of juvenile herons occurred, but also lower in a number of locations where no herons where present. Predation could only take place in shallow habitats (<50cm) and therefore, lower predation pressure was thought to take place in lakes and large rivers, where herons can only fish on the banks. Table 7.4 summarises published data on eel consumption by cormorants in Europe.

Evidently the composition of the cormorant diet varies much both in time and space. The availability of eels is low in winter and there is a decrease in the proportion of eels found in the Danish data, which may reflect the decline in the eel abundance.

The food intake of cormorants is approximately 400-500 g/day (Keller and Visser 1999). Present estimate of the European breeding population is 250 –300 000 pairs. The range of total eel consumption by cormorants can then be estimated to be somewhere between 1 800 ton/y (2% of diet whole year) and 9 000 ton/y (20% of diet half of year).

The size distribution of eels in cormorant diet differs from that in the fishery. A large part of the eels eaten by cormorants will be small, mean weight 100 g/eel, so the consumed weight corresponds to approximately 2 times this amount if the same eels were exploited in the fishery.

Seal predation is probably insignificant. Harbour seal (Phoca vitulina) diet has been studied (Härkönen 1987, Lunneryd 2001) and it was found that eel remains were absent in the scat and that seals presented with a choice of different fishes preferred other species than eels. It is known that both harbour seals and grey seals (Halichoerus grypus) can specialise in taking eels from fyke nets, and in this case they evidently can develop a taste for eels. This will however in practise be part of the fishery exploitation.

Whales are the dominant mammalian fish consumers. A review has been made of whale diets in the NE Atlantic by ICES (1999). No eel remains were found in the whale species studied, except for harbour porpoise (Phocoena phocoena) (Bjørge 1991). This species feed on the bottom and occurs regularly close to the coast. In the Norwegian study eel remains were found in 0.75 % of the samples.

The harbour porpoise population in the southern Irish Sea, the Channel, North Sea, Skagerrak and Kattegat is approximately 350 000 animals (Hammond et al 1995). The daily fish consumption is approximately 5 kg (ref xx). If the average component of eel is 0.5 % this means a total of 32 000 ton/y. As with bird predation it is likely that there is a seasonal variation with a negligible eel consumption in the winter. There is also just a partial overlap of the eel and harbour porpoise distribution ranges. As a tentative guess at any one time 25 % of the porpoises are so close to the shore that they are likely to find eels. With those assumptions the harbour porpoise consumption becomes 4 000 ton/y.

7.5 Restocking

Stocking of eels is traditionally a major ingredient of eel management in northern Europe. Section 2.3 discusses the trend in re-stocking over the past decades. Restocking as a management measure is different in principle from the ones discussed above, including fishery regulations. Change in habitat access and a decrease of anthropogenic or other causes of mortality can only move the escapement in the direction of the maximum found in pristine conditions. In principle, restocking can reach beyond this if otherwise under-colonised areas are used to full capacity.

There is an asymmetry in the advection of eel larvae to Europe. The east Atlantic current branches at the continental shelf Southwest of Ireland. A northerly branch carries eels north of Scotland into the North Sea and continues to the Skagerrak and Baltic. The other branch divides further and carries the eel larvae south to the Bay of Biscay and through the Channel into the southern North Sea.

37 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Those two transport routes divide the distribution range of the European eel into a northern and a southern part with fundamentally different conditions. The southern part is characterised by a comparatively small total freshwater habitat and a large inflow of glass eels. This is the region where glass-eel fishery is important. In the northern part of the range more than 75% of the freshwater and brackish water areas are situated and at present those waters to a large extent are under-utilised by eels.

A redistribution of glass eels from south to north already exist but there is scope for a large increase. Moriarty and Dekker (1997) have made an analysis of the possibilities and find that the whole glass eel catch can be used for stocking without overstocking the available productive eel habitats. They find that this should increase the eel yield by approximately 60000 ton/y.

One argument against redistribution of glass eel is the danger of damaging a possible genetic diversity in the eel population. Traditionally the hypothesis has been that the 3 year oceanic phase of the leptocephalus transport should guarantee a thorough mixing and decoupling between parent origin and the arrival place of the juvenile. This assumption has been questioned by DNA studies that indicate a north-south gradient in genetic markers (Wirth and Bernatchez 2001). There is an ongoing Belgian-Swedish study, based on a more extensive geographic range of samples, to verify these findings. So far the results point to a very small but significant genetic variation if seen over the whole spatial scale, but the possible implication for management is unclear. The sex distribution in different parts of the range of the eel also makes it difficult to understand how a north-south genetic difference could be maintained. The whole Baltic drainage area is effectively a unisex area with essentially only female eels. Either this should be a genetic cul de sac or the genome must be mixed each generation.

Another argument is historical: maximum restocking levels have been reached in most countries in the 1970s or early 1980s. This coincided with a maximum in natural recruitment. At the same time, a general decline in commercial yield was observed in most countries, which has continued ever since. Apparently, the high natural immigration in combination with the re-stocking added have never resulted in a considerable rise in yield. It is therefore unlikely that renewed boosting of recruitment to 1960s-1970s levels by artificial restocking will have an adequate effect on the continental production.

At the bottom line, local examples have proven the effect of re-stocking beyond doubt, while stock-wide effects seem to have been negligible and potentially carry a genetic risk.

7.6 Conclusions on exploitation vs. other measures

In spite of the very uncertain data available for estimates of the importance of the non-fishery related factors it seems clear that they are of a magnitude that makes them a viable alternative for eel restoration and escapement targets (Table 7.6).

In the short-term, installation of upstream eel ladders is probably the easiest achievable.

Somewhere in the Alsace An eel tried to climb up a pass But halfway to the top He came to a stop When he ran out of artificial grass! RR

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 38

8 POST-EVALUATION - DEVELOP DATA-RICH AND DATA-SPARSE PROCEDURES FOR EVALUATING THE EFFICACY OF MANAGEMENT MEASURES

8.1 Introduction

It is generally assumed, at present, that the European eel population is a single biological stock. However, since the spawning process is still largely unsolved, it will be prudent to aim for sustainable management schemes in most (if not all) catchments, and practical management actions will be applied within catchments (to "management stocks"). Feunteun (2002) describes the data requirements and modelling approaches to defining stock restoration targets and evaluating compliance at local, catchment, regional and European wide scales. Though it is unlikely that we have sufficient data for any eel management unit to make reliable analytical assessments for establishing reference points against which to judge stock status, a %SPR of 30 % as target for silver eel escapement has been adopted (ICES 2002). This could be used as the management goal, for which we need to define management targets (or their proxies) that all regions can work towards achieving (possibly at a catchment level). Consequently, methods for evaluating the effect (positive or negative) of implemented measures or other perturbations on the production of silver eels are required.

Eel is a slow growing fish. Consequently, a delay of up to 5 - 20 years between the implementation of a management measure (on glass eels, for example) and its effect becoming manifest in silver eel escapement is to be expected. There is, therefore, a need to identify means of monitoring the potential effects on a shorter time scale, in order to provide timely advice on the adequacy of management actions.

This section aims to present means ("indicators") of monitoring population parameters at relevant life stages and to present these in terms of data needs / availability and the time scale within which an evaluation can be carried out.

8.2 Development of Procedures

It is necessary to identify indicators that signify change and which can eventually be evaluated against management targets or proxies on the status of the stock. It is not necessary, however, that the indicators be defined in terms of silver eel escapement (%SPR). These could, for example, be:

• Escapement of elvers (recruitment to the stock) • Size or age structure in the yellow eel population - by sex • Density - by life stage, sex and maturity • Sex ratio, at size, at age, in yellow eel population (especially at silvering) • Age at silvering - by sex • Growth rate • Escapement of silver eels, as proportion of potential, or actual production • Yield-per-recruit (YPR) • Spawner-per-recruit (SPR) • Quality of spawners in terms of contamination/ parasite loads etc

Though management actions (in response to targets or their proxies) will probably be catchment specific, some indicators may have variation or gradients within a catchment that need to be taken into account in evaluating the results of monitoring.

Preferably, indicators should be easy to implement and clearly indicative of the state of the fishery/population.

Changes in indictors can be evaluated in a relative sense (in order to judge the direction and magnitude of a management measure's effect) or in absolute terms (e.g. to evaluate compliance with a 30%SPR target).

8.3 Timeliness

There will be a requirement to consider the impact of management or other perturbations (positive or negative) on eel populations through indicators that are measurable at different time scales. Some management actions are adaptive, such that rapid feedback (within one year) of their efficacy is required to adjust the intensity of regulation in order to justify the measures against its effect. In other cases, it may be sufficient (or unavoidable) to use indicators that are measurable only in the medium (e.g. 3-10 years) or longer term.

39 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

For example, an indicator of glass eel escapement could be used to evaluate changes due to regulations on a glass eel fishery and these may be detectable in the summer season following immigration. Significant changes in an indicator of silver eel production (e.g. due to habitat modifications; elver recruitment failure, yellow eel exploitation) may only be detectable in the long-term, though short-term change may be evident if a result of a substantial environmental influence, such as pollution. Changes in YPR are likely to be detectable in the medium-term.

In summary, short-term indications are possible where the catch of a life stage is dominated by currently exploited cohorts; medium-term indications where the catch is dominated by the cohorts that are present in the population; and long-term indications rely on the recruitment of new cohorts (does this help?)

Table 8.3. The likely time lag between a management action on each life stage and the availability of an indicator that will enable the effect of any change to be evaluated.

Effect Local Management Stock Young Old Whole Glass Yellow Yellow Silver Population Glass Young Yellow

Stock Old Yellow

Local Management Silver Management change

Key short medium long not possible

8.4 Feasibility of detecting a change in the indicator, and its implications

The possibility that the effect of a management measure can be detected by a change in an indicator will depend on the availability of relevant data (or the likelihood that data can be obtained) and the "natural" variability of the indicator under status quo conditions. For example, changes in mean size of males / females or sex ratio may be apparent and useable only if there is a significantly large effect on the population. In contrast, a cessation of silver eel entrapment should be immediately evident through an indicator linking silver eel production with silver eel escapement from the catchment.

It will be necessary to determine what degree of change is considered to be significant in terms of the performance of a local/regional fishery or stock, though this cannot be applied at present to global eel sustainability.

In many circumstances, it may be difficult to distinguish changes in indicators due to management actions from those due to other anthropogenic or environmental impacts (including habitat destruction or restoration). For instance, changes in age structure may indicate the effects of changes in the level and pattern of fishing mortality, but they could also result from recruitment failure. As a consequence, it is necessary to take account of other potential causes of change in eel populations. These include the relationship between eel population structure and production in relation to the rest of the ecological community; factors influencing the eel stock may have a direct or indirect impact on other parts of the community, and eel production may be directly or indirectly affected by perturbations in the community. A current example is the impact on eel population densities and fishery yield due to increased cormorant predation in many parts of Europe.

Note also that there may be constraints on potential changes (e.g. high natural mortality between glass eel and elver stages) that obscure those resulting from management regulations (i.e. bottleneck effect).

8.5 Methods for collecting and analysing relevant data

We have not attempted to provide an exhaustive list of methodologies or indicators, but provide examples for each life history stage that may serve as a basis for their development.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 40

8.5.1 Glass Eel

Glass eel escapement (recruitment to the river) could be monitored by proportional pigmentation stage (which provides an indication of the scale of fishing mortality on glass eels, and may be independent of the actual level of recruitment to the estuary), glass eel fishery CPUE (taking into account the efficiency of gears), elver counts, fishery-independent survey of early yellow eel ages, etc.

Case study. Knights et al (2001) examined criteria for developing a monitoring strategy that could be used to advise management. The study focused on the glass eel fishery on the lower River Severn - a major source of glass eel recruitment within England – with the aim of ascertaining the level of sampling needed to detect changes in density relative to some management reference level. Specifically, they evaluated the number of sample sites required to reliably detect changes in density between repeat surveys, taking into account that real changes may be obscured by natural variability (spatial and/or temporal). The results demonstrated that 26 sample sites were needed to detect a 50% change between two consecutive surveys. However, the authors noted natural variation, even with small sampling intervals, and recommended that similar evaluations of other river systems will improve understanding of the sensitivity of perturbations as well as the differences among regions.

There is evidence that a combination of management and habitat improvement can have a positive impact on a local eel stock within a watershed. Briand et al (in press) evaluated the effects of installing an eel ladder and implementing fishery regulations on the Vilaine watershed eel stock. During the winter, glass eel are harvested in the estuary as they cumulate near the Arzal dam, which prevents their passage to upper reaches of the watershed. In 1996, an eel ladder was constructed to improve glass eel escapement and increased eel density in riverine areas. Limits were also imposed on the length of the fishing season. Electrofishing and eel ladder counts were used to evaluate the efficacy of the ladder and management action. The results showed that the eel ladder was effective in allowing glass eels cross the dam, there was a marked increase in the density of young eels, and there was a transition from domination of age 3 and 4+ to domination by age 1 in the middle and downstream sectors of the watershed. The scarcity of age 1 in these areas prior to the changes may reflect that only the larger/older eel had the ability to cross the dam prior to the installation of the ladder.

8.5.2 Yellow Eel

Through a sampling programme of commercial fishing or survey catches, age structure could be evaluated by measuring relative abundance of the size/age group of interest (e.g. look at age and size composition in population and density, numbers-at-size/age by hectare) (examples required). It may be less useful to measure size-at-age, growth, or sex ratios, given the spatial and temporal variability in these parameters and the slow response time to perturbations.

8.5.3 Silver Eel

Production/escapement of silver eel may be monitored by identifying the proportion of the length group that is likely to become silver (using pre-emigration diagnostics for potential silvering), and use estimates of population density to predict how many silver eels can be expected per hectare of given habitat. Then, at the time of migration, look at the proportion escaping (e.g. through fishery or other monitoring, mark and recapture, e.g.). This may, of course, serve as a proxy for the management target.

Table 8.5 illustrates the potential feasibility of achieving an indicator (of the effect of a management action on each life stage and a resultant change in eel population or yield).

41 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Table 8.5. The potential feasibility of achieving an indicator (of the effect of a management action on each life stage and a resultant change in eel population or yield). Local Management Stock Young Old Whole Glass Yellow Yellow Silver Population Glass

Young Yellow

Stock Old Yellow

Local Management Silver

Key high low not possible

8.6 Conclusions on post-evaluation

The eel population as a whole has experienced a slow and long-lasting decline over the whole distribution area. Management approaches for maintaining a sustainable eel resource through stock-wide management targets should, in practice, be catchment specific, and need to be implemented locally. A number of measurable parameters could serve as potential indicators of the effectiveness of management measures, but there are obvious advantages in identifying those indicators that can be reliably estimated, before implementing a management programme. Indicators should be chosen that reflect changes resulting from the proposed management action (in relation to targets or their proxies) and are useful for determining the status relative to the overall management goal. The most suitable properties of indicators for monitoring population parameters are discussed, and examples of post-evaluation methods have been provided. Post- evaluation (via indicators) at the local (catchment) scale will evaluate the efficacy of management within a catchment and, combined with results from other catchments, help with assessing the influence of management on the overall condition of the stock. Further work is needed to develop guidelines for identifying and evaluating appropriate indicators.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 42

9 CONTAMINATION AND EELS

9.1 Introduction

Eels are particularly sensitive to bioaccumulation of contaminants like chlorobiphenyls, organochlorine pesticides and heavy metals. They are long-lived, benthic fish, feeding on insect larvae, worms, Crustacea, snails, mussels and fish, in particular small bottom dwelling species (Tesch 1977). This results in high bioaccumulation of toxic residues. Most of these contaminants are lipophylic. As eels have high lipid concentrations compared to other fish, contaminant concentrations may attain very high levels.

9.2 Status

An overview of the potential effects of these contaminants on spawning success and larval survival was made by Robinet and Feunteun (2002). Although these effects are poorly documented in eels, many studies are known from other fish species. Eggs concentrate lipophylic contaminants, and these contaminants may have a negative impact on ovarian recruitment. Reproductive performance and spawning success is reduced. Negative impacts on several aspects of the reproductive physiology of fish were documented. These findings led to the belief that contamination of eels by a number of chemical compounds might be an important factor in the decline of the Atlantic eel species by interacting on the reproduction.

Contamination of various fish species (none on eels) by various lipophylic and non lipophylic compounds have been shown to effect key life functions as muscular activity, lipid storage, fertility, sex differentiation, larval survival and growth, etc. at very low concentrations. However, most of the studies on this area were conducted in vitro and no studies relate contaminant concentrations in the water, pathways to the cells, body concentration and physiological disturbance. All the contamination levels at the (inter)national levels are fixed to ensure human health standards but they are not based on ecotoxicological criteria. Studies are needed to evaluate effects of contamination by different compounds on reproduction and to fix maximal acceptable concentration levels on sound ecotoxicological basis.

9.3 Legislation

Recently, international and, in some cases, national legislation has been implemented on the maximum level for contaminants in foodstuffs. These regulations, induced by human health considerations, might in the short-term have a considerable impact on fisheries and stock management (by i.e. closing the fisheries).

This unintentional outcome is in contrast with the fact that for many years eel managers have attempted unsuccessfully to set up international eel management plans for recovering the stock, including a restriction of the fisheries.

The new EC regulation (EC 2001a) sets maximum levels for certain contaminants (polychloordibenzo-p-dioxines and polychloordibenzofuranes) in foodstuffs allowing for a maximum of 4 pg PCDD/F-TEQ /g fresh weight for muscle meat of fish and came into force on 1 July 2002.

In Sweden, PCDD/F-TEQ values in eels (both yellow and silver eels) of 9 sites in fresh water lakes, in the Baltic Sea and along the west Coast, were all lower than the maximum level (http://www.slv.se). Recent results of analysis of wild eels in the Netherlands, however, showed that 7 sites of 39 (18%) exceeded the allowable levels (Van Leeuwen et al. 2002). In Lake IJsselmeer, eels from 2 of 6 sites were higher than this maximum level.

It is likely that, in near future, these findings will lead to a restriction of the eel fisheries in some areas. The potential stock-wide impact is difficult to predict.

In Belgium, by means of a new Royal decree, the government adopted a new PCB standard for fish and fisheries products. According to this, the sum of the seven marker PCBs (polychloorbiphenyls PCB28, PCB52, PCB101, PCB118, PCB138, PCB153 en PCB180) may not exceed 75 µg/kg on product basis (Belgisch Staatsblad 2002a). In Flanders, a monitoring network for measuring PCB concentrations in eels has been set up in recent years. According to the results of the analysis of 1057 eels from 244 locations (Figure 9.3) in Flemish canals, rivers and closed water systems, this standard is exceeded at 80% of the sites. In some cases, mean concentrations of the sum of the seven marker PCB exceed 6700 µg/kg on product basis (Goemans and Belpaire 2002). As a consequence of these results, the Flemish Minister for Environment has banned the taking of eels from public inland waters in Flanders until 2006 (Belgisch Staatsblad 2002b). Moreover, the use of a number of fisheries devices specialised for catching eels like fykes and liftnets is now prohibited in most of the Flemish rivers and canals (Belgisch Staatsblad 2002c).

43 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Indirect effects of such bans due to contamination are likely to increase the production of silver eels by reducing fishing mortality. However, studies are needed to quantify these unintentional effects, and must also take account of the low quality of contaminated eels as potential spawners.

Also in other countries PCB concentration in eels were measured and are known to be higher then in other species (e.g. in France on the Gironde (Goursolle 2002) and in the Netherlands (Leonards et al. 2000)) giving reasons for taking precautionary measures. In general PCB concentrations showed decreasing trends in a number of ecosystems. However new chemical compounds like polybromebyphenols and hormonal disrupters are currently increasing in most biota all over Europe.

8000

7000

6000

5000

4000

3000

Concentration (µg/kg body weight) 2000

1000 Belgian consumption standard = 75 µg/kg 0 Site

Figure 9.3 Mean PCB concentrations in eels in Flanders (µg/kg on product basis for the sum of the seven marker PCBs in eels from 244 sites, 1994-2000). Legal maximum for fisheries products is 75 µg/kg (Institute for Forestry and Game Management, Flanders).

9.4 Conclusions on contamination

National and supranational legislations enforce evaluation of contaminant accumulation of eels within the EC member states. The unintentional effect of these human food safety regulations on eel stock and commercial exploitation should be evaluated at a population-wide basis.

As deterioration in the quality of eel spawners (possibly comparable to the potential effects of parasitism in the swimbladder, changes in ocean currents and variations in the oceanic secondary productivity (Robinet and Feunteun 2002). Up to now, no such study is known, despite possible effects on human health and on the breeding potential of various eel sub-populations.

Considering the geographical variation in use of various contaminants over countries and catchments and temporal variations (new contaminants used in some areas, interdiction of old ones, …), monitoring and analysis have to take into account these evolutions on a catchment basis and adapt to changes over time both in contaminant uses and in (inter)national legislations.

An eel from Nantes, on the Loire Thought the Sargasso was much too far, So she stayed in the river, Got PCBs in her liver And was eaten in St Nazaire. RR

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 44

10 CONCLUSIONS AND RECOMMENDATIONS

10.1 Conclusions

Review of the available information on the status of the stock and fisheries of the European eel supports the view that the population as a whole is in decline in most of the distribution area, that the stock is outside safe biological limits and that current fisheries are not sustainable. Recruitment is at a historical minimum and most recent observations indicate the decline continues. Evidence has been given that anthropogenic factors (e.g. exploitation, habitat loss, predation, contamination and transfer of parasites and diseases) as well as natural processes (e.g. climate change) have contributed to the decline. Measures aimed at recovery of the stock are well known and may include control of exploitation, restocking of recruits or restoration of habitats (including access to and from).

The continental population extends throughout Europe and northern Africa and fisheries are scattered over many large and small water bodies. Management at the local level has not adequately addressed the global decline of the stock, and no co-ordinated stock-wide management framework has been set up. The continuation of the decline demonstrated by most recent data makes the compilation and implementation of an international stock recovery plan of a growing urgency.

Current conclusions on trends the stock and its status are based on national monitoring programmes, several of which have come under the increasing pressures of budget cuts. A programme for monitoring of fisheries and stock recovery is a fundamental component of the management plan indicated above. However, pending the compilation and implementation of a recovery plan, it is necessary that monitoring of recruitment, stocks, fisheries and escapement be sustained at least at current levels. To this end, it is recommended that countries report annually on trends in local populations and fisheries to the Working Group. These national reports should comprise data on recruitment, fishing effort, landings and where possible also on non-exploitation related factors, e.g. restocking, habitat, pollution (see section 2.6 for details).

Management options aiming at protecting the stock refer primarily to whole-stock conservation limits that need to be translated into appropriate local-system targets. Local management will depend on the available knowledge from the area. In order to derive local management schemes, further development of research on several aspects of the biology and exploitation of eels will be required. In this report, some of these issues have been explored, notably density dependent processes, loss of habitat, derivation of reference concepts and options for post-evaluation of the effect of management measures on the stock. Further development and strengthening of the international co-operation and co- ordination are recommended.

10.2 Recommendations

The ICES/EIFAC Working Group on Eels at its 2002 session in Nantes (France) recommends that:

• A recovery plan for the eel stock is compiled and implemented as a matter of utmost urgency and that fishing and other anthropogenic mortality be reduced to the lowest possible level until such a plan is agreed upon and implemented.

• Monitoring of recruitment, stocks, fisheries and escapement should at least be sustained at recent levels, whilst a stock recovery plan - including a comprehensive monitoring and research programme - is agreed upon and implemented.

• Member countries report annually on trends in their local populations and fisheries to the Working Group.

45 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

11 LITERATURE REFERENCES

Acou, A. Feunteun E., Laffaille P., Legault A. 2000 Catadromous migration of European eel (Anguilla anguilla, L.) in dammed catchments. Verh. Int. Ver. Limnol N° 27

Acou, A. 1999

Acou, A. 2001

Adam G. 1997 L'anguille européenne (Anguilla anguilla L. 1758) : dynamique de la sous-population du lac de Grand- Lieu en relation avec les facteurs environnementaux et anthropiques. TH, 353 pp.

Adam, B. 2000 Abstiegsanlage oder Frühwarnsystem? Neue Erkenntnisse zur Gewährleistung der Aalabwanderung. 14. SVK-Fischereitagung. 1.-2.3.200 in Künzell b. Fulda.

Adam, B., Schwevers, U. & Dupont, U 2002 Beiträge zum Schutz abwandernder Fische – Verhaltensbeobachtungen in einem Modellgerinne. Bibliothek Natur & Wissenschaft, Solingen 16: 64 pp.

Adam, G. 1997 L’anguille européene (Anguilla anguilla): dynamique de la sous-population du lac de Grand-Lieu en relation avec les facteurs environmentaux et anthropiques. PhD thesis. U Paul Sabatier Tolouse III

Anonymous 2000 Gestion de l'anguille (management of eel) / Plan de gestion 2001-2005 des poissons migrateurs (management plan for migratory fishes in Brittany). Rennes: Préfecture de la région de Bretagne. 12-22 pp.

Aprahamian, M. W. 2000 The growth rate of eel in tributaries of the lower River Severn, England, and its relationship with stock size. Journal of Fish Biology 56: 223-227.

Ask L. and Erichsen L. 1976 Blankålsmärkningar vid svenska Östersjökusten 1941-1968. Meddelande från Havsfiskelaboratoriet Lysekil 199, 1-117.

ASMFC 2000 Interstate fishery management plan for American eel. Fishery Management Report No. 36.

Baisez A. 2001 Optimisation des suivis des indices d'abondances et des structures de taille de l'anguille européenne (Anguilla anguilla, L.) dans un marais endigué de la côte Atlantique : Relation espèce - habitat., Université de Toulouse III - CEMAGREF, Toulouse pp.

Baisez A., Rigaud C., Feunteun E. 2000 Hétérogénéité de répartition de l’anguille européenne (A. anguilla) dans un marais endigué de la côte atlantique (marais breton). Cybium.34 (3) : 23-32

Baudry D. 1999 Evaluation des zones de frayèes et de production en juvéniles pour la Grande Alose (Alosa alosa L.) au niveau de l'Adour. Rapport Institution Adour-CSP-IFREMER, 39 pages + annexes.

Belgisch Staatsblad 2002a. 16 April 2002 - Koninklijk besluit tot wijziging van het koninklijk besluit van 19 mei 2000 tot vaststelling van maximale gehaltes aan dioxines en polygechloreerde bifenylen in sommige voedingsmiddelen (6 March 2002)

Belgisch Staatsblad 2002b. 25 May 2002 - Ministerieel besluit houdende een tijdelijk meeneemverbod van paling in alle openbare wateren en een tijdelijk meeneemverbod van alle vissen op bepaalde openbare wateren 19 April 2002)

Belgisch Staatsblad 2002c. 25 May 2002 - Ministerieel besluit houdende een tijdelijk verbod op het gebruik van de palingfuik en het kruisnet in de grensscheidende Maas en de niet-bevaarbare waterlopen en kanalen in de provincies Oost-Vlaanderen en West-Vlaanderen 19 April 2002)

Belpaire, C., De Charleroy, D., Coeck, J., Janssens, L. and Monden, S. 2002. Flemish plans for hydropower turbines : additional loss for eel spawning stock? ICES/EIFAC - Working Group on Eel (Nantes, 2-6 September 2002)

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 46

Belpaire, C., van Vlasselaer, L., Podoor, N. and Ollevier, F. 1989 Raising of glasseels in ponds [in Dutch]. Report Zool. Inst. Catholic University Leuven Belgium 104p.

Benelux 1996 Decree on the free migration of fish species in the hydrographic river basins of Benelux countries. Committee of Ministers of the Benelux Economic Union, Document M (96) 5, The Hague, April 26th 1996

Berg, R 1988 Gutachterliche Stellungnahme zu Fischschäden durch den Betrieb der Wasserkraftanlage „Am letzten Heller“. Gutachten Fischereiforschungsstelle Baden-Würtemberg: 34 pp.

Berg, R. 1985 Turbinenbedingte Schäden an Fischen – Bericht über Versuche am Laufkraftwerk Neckarzimmern. Gutachten Fischereiforschungsstelle Baden-Würtemberg: 25 pp.

Berg, R. 1993 Untersuchung einer Fischscheucheinrichtung am Kraftwerk Neckarzimmern. Gutachten Fischereiforschungsstelle Baden-Würtemberg: 20 pp.

Berg, R. 1994 Untersuchung einer Fischscheucheinrichtung am Kraftwerk Guttenbach (Neckar). Gutachten Fischereiforschungsstelle Baden-Würtemberg: 25 pp.

Bjørge A., Aarefjord H., Kaarstad S., Kleivane L., Øien N. 1991. Harbour porpoise (Phocena phocena) in Norwegian waters. ICES CM 1991/N:16

Briand, C., F. Denis, G. Fontenelle, and E. Feunteun. Effect of re-opening of a migratory axis for eel at a watershed scale. American Fisheries Society Symposium presentation.

Briand 2001 Gestion du stock d’anguille dans le bassin versant de l’Aulne. Regional report COGEPOMI 2001.

Briand C., Fatin D., Feunteun E. & Fontenelle G. In press Estuarine and fluvial recruitment of European glass eel in a fished Atlantic estuary. Fisheries Management and Ecology.

Briand C., Fatin D., Fontenelle G. & Feunteun E. 2000. Effect of re-opening of a migratory axis for eel at a watershed scale (Vilaine river, Southern Brittany). American Fisheries Society Symposium

Briand C., Mounaix, B., Fatin D., Feunteun E. in presStock contribution of springtime and Autumn glass eel recruitments: results based on otolith structure. Bulletin français de pêche et pisciculture.

Briand, C., F. Denis, G. Fontenelle, and E. Feunteun in press Effect of re-opening of a migratory axis for eel at a watershed scale. Which journal?

Bunge Th., D.Dirbach, B.Dreher, K.Fritz, O.Lell, B.Rechenberg, J.Rechenberg, E.Schmitz, S.Schwermer, m.Steinhauer, C.Steudte &T.Voigt, 2001 Wasserkraftanlagen als erneuerbare Energiequelle. Rechtliche und okologische Aspekte. Umweltbundesamt, Berlin, 91 p.

Busch, W.D.N., S.J. Lary, C.M. Castilione and R.P. McDonald. 1998 Distribution and availability of Atlantic Coast Freshwater Habitats for American Eel (Anguilla rostrata). Adminstrative Report #98-2. USFWS, Amherst, NY 28pp.

Casamajor (de), M.-N., Bru NB. Et Prouzet P. 1999 Influence de la luminosité nocturne et de la turbidité sur la comportement vertical de migration de la civelle d’anguille (Anguilla anguilla L.) dans l’estuaire de l’Adour. Bull. Fr. Pêche et Pisci., 353, 327 – 347.

Castelnaud G. 2002 Caractéristiques de la pêcherie civellière du golfe de Gascogne.Contrat européen n° 99/023 Cemagref, Groupement de Bordeaux, 32 p.

Castelnaud G., Girardin M., Rochard E. 2001 Surveillance halieutique de l'estuaire de la Gironde : suivi des captures 1999 et étude de la faune circulante 2000. Etude Cemagref Groupement de Bordeaux, EDF, n° 71, 185 p.

Changeux T. 2002 L’effort de pêche ciblant l’anguille dans les eaux libres Françaises. Rapport du Conseil Supérieur de la Pêche, Paris, 6 p.

47 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Changeux T., Gianneti M.-L., Rancon J. 2002. Evaluation des captures d’anguilles par les membres d’AAPPMA dans le bassin de la Loire, cas du département de Loire-Atlantique, enquête sur l’année 2000. Rapport Conseil Supérieur de la Pêche - LOGRAMI, 8 p.

Dahl J. 1983 Some observation on the ascent of young eels at the Tange Power Dam, river Gudena. FAO/EIFAC, Working Party on Eel, Stockholm 1982, 19-22.

Deelder C. L. 1952 On the migration of the elver (Anguilla vulgaris Turt.) at sea. J. Cons. Int. Expl. Mer, 18, 187-218.

Deelder C. L. 1958 On the behaviour of elvers (Anguilla vulgaris. Turt.) migrating from sea into fresh water. J. Cons. Int. Explor. Mer. 24 , 135-146.

Dekker W. [ed.] 2002 Monitoring of glass eel recruitment. Report C007/02-WD, Netherlands Institute of Fisheries Research, IJmuiden, 256 pp.

Dekker W. 2000a The fractal geometry of the European eel stock. ICES Journal of Marine Science 57:938-947

Dekker W. 2000b A Procrustean assessment of the European eel stock. ICES Journal of Marine Science 57:109-121

Dekker W. 2000c Impact of yellow eel exploitation on spawner production in Lake IJsselmeer, the Netherlands. Dana 12:17-32

Dekker W. 1991 Assessment of the historical downfall of the IJsselmeer fisheries using anonymous inquiries for effort data. P 233-240 in: Cowx, I.G., 1991, Catch Effort sampling strategies, their application in freshwater management. Fishing News Books, Oxford. 420 pp.

Desaunay D., Guerault D., Beillois. 1987 Dynamique de la migration anadrome de la civelle (Anguilla anguilla) dans l’estuaire de la Loire : rôle des facteurs climatqiues vis à vis de la pêche et du recrutement. C.I.E.M. Comité des poissons anadromes et catadromes C.M. 1987 / M18.

Dönni, W., Maier, K.-J. & Vicenti, H. 2001 Bestandsentwicklung des Aals (Anguilla anguilla) im Hochrhein. Mitt. zur Fischerei, BUWAL, Bern, 27, 99 pp.

EC 2001a Council regulation (EC) No 2375/2001 of 29 November 2001 amending Commision regulation (EC) No 466/2001 setting maximum levels for certain contaminants in foodstuffs. Official journal of the European Communities, 6.12.2001, L 321/1

EC 2001b Directive 2001/77/EC of the European parliament and of the council of 27 September 2001 on the promotion of electricity produced from renewable energy sources in the internal electricity market. Official Journal of the European Communities 27.10.2001 L 283/33

European Commission Fisheries Directorate General 1999 Forward Study of Community Aquaculture, MacAlister Elliott and Parners Ltd

Eybert M.C, J.Y. Bernard, P. Constant, E. Feunteun, J. Hédin, S. Questiau 1998 Marsh meadow restoration in the Brière marsh (France) : evolution of flora, birds and fishes. Gibier Faune Sauvage. 15 : 999-1016

Faveri N. de 2002 Zone humides des functions multiples. Revue de l’Agence de l’Eau, 84:3-8.

Feunteun & Boisneau, xxx , silver tagging

Feunteun E. 2001 Movements in inland habitats : adaptive value for anguillid eels. Advances in eel biology, Tokyo 28- 30 septembre 2001.

Feunteun E., A. Acou, J. Guillouët, P. Laffaille, A. Legault 1998 Spatial distribution of an eel population (Anguilla anguilla L.) in a small coastal catchment of northern Brittany (France). Consequences of hydraulic works. Bull. Fr. Pêche et Piscic. 349 : 129-139

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 48

Feunteun E., Acou A., Laffaille P., Legault A. 2000 The European Eel (Anguilla anguilla, L.): prediction of spawner escapement from continental population parameters, Can. J. Fish. Aquat. Sci. 57 : 1627-1635.

Feunteun E., C. Rigaud, P. Elie & J.C. Lefeuvre 1992 Le marais doux endigué de Bourgneuf Machecoul (Pays de Loire). Premiers éléments de connaissance de la faune piscicole. Relations ichtyofaune-habitat et problèmes majeurs de gestion. Rev. Sci. Eau. 5: 509-528

Feunteun E., C. Rigaud, P. Elie et J.C. Lefeuvre 1999 Les peuplements piscicoles des marais littoraux de l’ouest de la France : un patrimoine à gérer ? Bull. Fr. Pêche et Piscic. 352 : 63-79.

Feunteun 1994 reclaimed salt marshes

Feunteun, E. 2002 Management and restoration of European eel population (A. anguilla): an impossible bargain. Ecological Engineering 18: 575 –591.

Feunteun, E. et al. 2000 L’anguille du Rhone aval. Direction region. L’environment Rhone-Alpes : 107 pp.

Feunteun, E., and Marion, L. 1994 Assessment of grey heron predation on fish communities : the case of the largest European colony. Hydrobiologia 279/280: 327-344.

Frost, W.E. 1950 The eel fisheries of the River Bann, Northern Ireland and observation on the age of silver eels. Journal du Counseil international pour l’Exploration de la Mer 16, 358-383.

Gascuel D. 1986 Flow carried and active swimming migration of the glass eel (Anguilla anguilla) in the tidal area of a small estuary on the French Atlantic coast. Hegol. Meeresunters., 40: 321-336.

Gascuel D., Feunteun E., Fontenelle G. 1995 Seasonal dynamics of estuarine migration in glass eels (Anguilla anguilla). Aquat. Living Resour., 1995, 8, 123-133.

Gascuel, D. 1986 low carried and active swimming migration of the glass eel (Anguilla anguilla) in the tidal area of a small estuary on the French Atlantic coast. Helgoländer Meeresunters 40:321-336.

Germonpre, F., Denayer, C., Belpaire, C. & Ollevier, F. 1994 nventarisatie van pomgemalen in het Vlaamse gewest en preliminair onderzoek naar de schade von diverse pomptypes op vissen na gedwongen blootstelling. Inst. voor bosbouw en wildbeheer & Kath. Univ. Leuven: 63 pp.

Goemans, G. and Belpaire, C. 2002 New Belgian standard for fish and fisheries products leads to catch and release obligation for eel (Anguilla anguilla) by fishermen in Flanders. Workshop on PCBs, Brno (Tchech Republic), 7- 10 May 2002

Goursolle, R. 2002. tude de la contamination en polychlorobiphényles le long des réseaux trophiques estuariens. Application à l'estuaire de la Gironde. Université Bordeaux, Laboratoire de physico-toxicochimie des systèmes naturels, 82 pages + annexes

Guérault D., Désaunay Y., Beillois P. & Grellier P. 1991 Etude des montées tardives de civelles et des conditions de colonistation des bassins versants de Loire et de Vilaine (Février à Juin 1990). IFREMER, Rapport DRV/RH/OCEAL, Nantes.

Guérault D., Prouzet P., Désaunay Y. & Beillois P. 1991 The recent evolution of the glass-eel immigration in three french atlantic estuaries. In Eifac working party on eel, pp. 18, Dublin (Ireland).

Guillouët J., Acou A., Mounaix B., Legault A. & Feunteun E. 2000 Etude de la restauration de la population d'anguille sur le Frémur : synthèse du suivi de 1995 à 1999. Rennes: Fish Pass. 120p pp.

Hald-Mortensen, P. 1995 Danske skarvers fødevalg i 1992-94 baseret på analyser af gylp fra 23 forskellige kolonier samt to overnatningspladser. Skov- og Naturstyrelsen, Miljø- og Energiministeriet.

49 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Hammond, P.S., Benke, H., Berggren, Borchers, D.L., Buckland, S.T., Collet, A., Heide-Jørgensen, M.P., Heimlich- Boran, S., Hiby, A.R., Leopold, M.F. & Øien, N. 1995 Distribution and abundance of the harbour porpoise and other small cetaceans in the North Sea and adjacent waters. Final Report to the European Commission under contract LIFE 92-2/UK/027. 242 pp.

Härkönen, T. 1987 Seasonal and regional variations in the feeding habits of the harbour seals (Phoca vitulina) in the Skagerrak and the Kattegat. J. Zool., Lond. 213:535-543

Holzner, M. 1999 Vermeidung von Fischschäden im Kraftwerksbereich. Schriftenreihe Bayr. Landesfischereiverb. 1: 224 pp. http://vismigratie.instnat.be/ Internetsite and database on the localisation and status of fish migration barriers in Flanders, Belgium. Ministery of the Flemish community. http://www.slv.se Website of the Swedish National Food Quality Board in Uppsala

Hvidsten N. A. 1983 Ascent of elvers (Anguilla anguilla) in the stream Imsa Norway. Report of the institute of freshwater research 62, 71-74.

Ibbotson, A., Smith, J., Scarlett, P. and Aprahamian, M. 2002 Colonisation of freshwater habitats by the European eel Anguilla anguilla. Freshwater Biology 47: 1696-1706.

ICES 1988 European Eel Assessment Working Group report, September 1987, International Council for the Exploration of the Sea, Copenhagen, Denmark.

ICES 1999 Report of the Working Group on Marine Mammal Population Dynamics and Trophic Interactions. ICES C.M. 1999/G:3

ICES 2000 Report of the EIFAC/ICES Working Group on Eels. ICES CM 2000/ACFM:6.

ICES 2001 ICES cooperative research report N° 246, Report of the ICES Advisory Committee on Fishery Management, 2001: 357-366.

ICES 2002 Report of the ICES/EIFAC working group on eels. ICES C.M. 2002/ACFM:03.

IKSR 2001 Rhein-Ministerkonferenz. Arbeitsplan bis 2005, 29. Januar 2001 in Strassburg. 41 p.

Jellyman, D. J. & P. R. Todd, 1998 Why are migrating male shortfinned eels (anguilla austrailis) in Lake Ellesmere, New Zealand, getting smaller and but not younger. Bulletin Français de la Péche et de la Pisciculture 349, 141- 152

Kamstra, A. 1999 Haltung, Fütterung und Aufzucht. In: Tesch, F.W. Der Aal (Chapman and Hall, London 1999).

Keller T., Visser H. 1999.Daily energy expenditure of great cormorants wintering at Lake Chimensee, southern Germany. Ardea 87:61-69

Kisker, G. 1930 Fische und Turbinen. Fischerei-Ztg. 33(22): 280 – 281.

Klein Breteler, J. G. P. 1992 Effect of provenance and density on growth and survival of glass eels Anguilla anguilla (L.) in mesocosm experiments. Irish Fisheries Investigations Series A (Freshwater) 36: 15-22.

Klein Breteler, J. G. P., Dekker, W. and Lammens, E. H. R. R. 1990 Growth and production of yellow eels and glass eels in ponds. Int. Revue ges. Hydrobiol. 75: 189-205.

Knights B., Bark A., Ball M., Winter E. & Dunn S. 2001 Eel and Elver stock in England and Wales - status and management options. Environment agency.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 50

Knights, B., A. Bark, M. Ball, F. Williams, E. Winter, and S. Dunn 2001 Eel and elver stocks in England and Wales – status and management options. Environmental Agency, Research and Development Technical Report W248. 294 pp.

Knösche, R. 2002 Ursachen des Aalrückganges und Möglichkeiten zur Förderung der Aalbestände in Mecklenburg- Vorpommern. Fischerei & Fischmarkt in Mecklenburg-Vorpommern 2 (3): 23 – 30.

Krueger, W. H. & K. Oliveira 1999 Evidence for environmental sex determination in the American eel, Anguilla rostrata. Environmental Biology of Fishes 55, 381-389.

Lainier, M. & F. Travade 1999 La Devalaison des migrateurs: problemes et dispositifs. Bull. Fr. Peche Piscic. 353/354: 181-210.

Larinier, M. & Dartiguelongue, J. 1989 La circulation des poissons migrateurs : le transit a travers les turbines des installations hydroelectriques. Bull. Français de la peche et de la pisciculture 312/313 (spec. Issue) : 53 pp.

Legault et a in press

Leonards, P.E.G., Lohman, M., de Wit, M.M., Booy, G., Brandsma, S.H. and de Boer, J. 2000 Actuele situatie van gechloreerde dioxines, furanen en polychloorbiphenylen in visserijproducten: quick and full-scan. RIVO Nederlands Instituut voor Visserijonderzoek, Ijmuiden, report CO34/00.

Lobón-Cerviá J., Utrilla C. G. & Rincón P. A. 1995 Variations in the population dynamics of the European eel Anguilla anguilla (L.) along the course of a Cantabrian river. Ecology of Freshwater Fish 4, 17-27.

Lunneryd, S.-G. 2001 Fish preference by the harbour seal (Phoca vitulina) with implications for the control of damages to fishing gear. ICES J. Mar. Sci. 58(4):824-829

Madsen, F.J. & Spärck, R. 1950 On the feeding habits of the Southern Cormorant in Denmark. Danish Rev. of Game Biology 1, 3: 45-75.

Marion L., Feunteun E, Rigaud C. and Carpentier A. 1998 Feeding strategies modification of Grey heron (Ardea cinerea L.) in response to a major change of the preyed fish community and diversity. SIL XXVII congress, August 8 – 14 1998, Dublin.

Marion L., Feunteun E., Carpentier A., Rigaud C. 2001 Feeding strategy modifications of Grey Herons (Ardea cinerea L.) in response to a major decline of preyed fish community biomass and diversity. Verh. Int. Ver. Limnol (Arch fur Hydrobiology). (sous presse)

Matthews, M, Evans, D., Rosell, R. Moriarty, C., and Marsh I. 2001 The Erne eel enhancement programme, Final Report. Northern Regional Fisheries Board, Ballyshannon, Co Donegal, Ireland. EU Peace and reconciliation Project No EU15.

McCarthy, T.K. and Cullen, P. 2000 Eel fishing in the River Shannon: Eel population changes, fishery management options and fishery conservation issues. A synthesis report on the River Shannon eel Management Programme 1992-2000. Electricity Supply Board, Dublin. 21pp

McCleave J.D.and Wippelhauser G.S. 1987 Behavioral Aspects of selective tidal stream transport in juvenile american eels. American Fisheries Society Symposium 1, 138-150.

McCleave, J.D.. 2001 Simulation of the Impact of Dams and Fishing Weirs on Reproductive Potential of Silver-Phase American Eels in the Kennebec River Basin, Maine. N. Amer. J. Fish. Mgt. 21(3): 592 – 605.

McGovern and McCarthy 1991 Elver migration in the river Corrib System, Western Ireland. EIFAC Working Party on Eels, Dublin, May 20-25, 1991

McGovern, P., and T. K. McCarthy. 1992. Elver migration into the river Corrib system, western Ireland. Irish Fisheries Investigations 36:25-31.

51 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Moriarty C [ed.] 1997 The European eel fishery in 1993 and 1994: First Report of a working group funded by the European Union Concerted Action AIR A94-1939. Fisheries Bulletin (Dublin) 14, 52 pp.

Moriarty, C and W Dekker 1997 Management of the European eel. Fisheries Bulletin (Dublin) 15, 110 pp.

Moriarty, C. 1986 Riverine migration of young eels (Anguilla anguilla). Fisheries Research 4:43-58.

Moriarty, C. 2001. Maintenance and development of the Corrib Catchment eel fishery. A report commissioned by the Western Regional Eel Fisherman's Association. 34pp.

MUNLV 2001 Das Wanderfischprogramm Nordrhein-Westfalen. Statusbericht zur ersten Programmphase 1998-2002. Ministerium fur die Umwelt und Naturschutz, Landwirtschaft und Verbraucherschutz des Landes Nordrhein- Westfalen, 40190 Dusseldorf, 112 p.

Muyres W. in prep Paper presented to the Internationale Kommission zum Schutz des Rheins.

Muyres W. 2002 Overzicht problematiek waterkrachtcentrales en ecologisch herstel trekvis in de riviersystemen. In prep.

Naismith, I. A., and B. Knights.1988 Migration of elvers and juvenile European eels, Anguilla anguilla L., in the river Thames. Journal of Fish Biology 33:161-175.

Parsons, J., K.Vickers & Y. Warden. 1977Relationship between elver recruitment and changes in the sex ratio of silver eels Anguilla anguilla L. migrating from Lough Neagh, Northern Ireleand. J. Fish Biol. 10, 211-229.

Pedersen M. I. and C. Dieperink 2000 Fishing mortality on silver eels (Anguilla anguilla (L.)), in Denmark. Dana 12: 77-82

Pollard et al 1987

Poole, W. R., J. D. Reynolds & C. Moriarty 1990 Observations on the silver eel migrations of the Burrishoole River system, Ireland 1959 to 1988. Int. Rev. gesamt. Hydrobiol. 75, 807-816.

Poole, W.R., and Reynolds, J.D. 1996. Growth rate and migration of the Eel Anguilla anguilla L., J Fish Biol. 48(4): 633-642

Poole, W.R., Reynolds, J.D., and Moriarty, C. 1990 Observations on silver eel migrations of the Birrishoole system, Ireland 1959 to 1988. Int. Revue Ges. Hydrobiolgie 75 (6) 807-815

Prignon, C., Micha, J.C & Gillet, A. 1998 Biological and environmental characteristics of fish passage at the tailfer dam on the Meuse river, Belgium. In : Fish migration and fish bypasses. in Jungwirth, M., Schmutz, S. & Weiss, S. (eds.). Fishing News Books, 69-84 - Turnpenny, A.W.H., 1998.

Prouzet P., Lissardy M., Morandeau G. et F.-X. Cuende 2001 La pêche maritime professionnelle dans l’estuaire de l’Adour en 2000 : Importance économique et caractéristiques des captures de poissons migrateurs amphihalins. Rapport IFREMER DRV/RST/RH/2001-08, 38 pages.

Prouzet P., Sanchez F., de Casamajor M.-N., Bru N. et R. Drouilhet 2000 Impact de la pratique du tamis poussé en zone maritime de l’Adour sur l’abondance des civelles et sur leur pêche. Rapport DRV/RH 09/2000_ Contrat PESCA et MAP, 83 pages.

Raben, K. von 1957 Über Turbinen und ihre schädliche Wirkung auf Fische. Z. Fischerei 6 N.F.: 181 – 182.

Rathke 1993

Rathke 1994

Rathke 1997

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 52

Robinet et al unpublished

Robinet, T. and Feunteun, E. 2002 Sublethal effects of exposure to chemical compounds : a cause for the decline in Atlantic eels? Ecotoxicology 11 : 265-277

Rosell R. 2000 Understanding the Dynamics of the Lough Neagh eel fishery. ICES working group paper.

Rosell R. 2001 Analysis of the Relationship between glass eel/elver input and eventual total catch in the Lough Neagh eel fishery. ICES working group paper.

Rossi, R., A. Carrieri, P. Franzoi, G. Cavallini & A. Gnes 1988 Eel population dynamics in the Comacchio Lagoons. Oebalia 14, 87-106

Sauvaget , B. 2001 Etude des stocks côtiers d’anguille en Bretagne, et évaluation du niveau d’exploitation du Golfe Morbihan, Institution d’amenagement de la Vilaine

Sers B., Meyer E. and Enderlein O. 1993 Sammanställning av fiskmärkningar utförda under åren 1980-85. Information från Sötvattenslaboratoriet, Drottningholm, no 4, 71p.

Späth, S. 1998 Charakterisierung turbinenbedingter Verletzungen an Mainfischen – dargestellt am Beispiel des Kraftwerkes Dettelbach. Diplomarb. Humboldt-Univ. Berlin, Landw.-Gärtn. Fak., Studiengang Fischwirtsch. und Gewässerbewirtsch.: 111 pp.

Svärdson G. 1976 The decline of the Baltic eel population. Inst. Freshw. Res. 55, 136-143

Svedäng, H 1999 Vital population statistics of the exploited eel stock on the Swedish west coast. Fisheries Research 40, 251 – 265.

Tesch, F.-W. 1977 The eel. Biology and management of anguillid eels. Chapman and Hall (London), 434p.

Teyssier F., Gayou F., Filszinger P., Galiay E., Lemoine A., Taillebois L. 2002 Plan de gestion des poisons migrateurs amphihalins Garonne – Dordogne – Charente – Leyre – Seudre. Situation et propositions. Conseil Superieur de la Peche, Groupe Anguille, 30+13 p.

Van Leeuwen, S.P.J., Traag, W.A., Hoogenboom, L.A.P., Booij, G., Lohman., Dao, Q.T. and de Boer, J. 2002 Dioxines, furanen en PCBs in aal. Onderzoek naar wilde aal, gekweekte aal, geïmporteerde en gerookte aal. RIVO Nederlands Instituut voor Visserijonderzoek, IJmuiden, report CO34/02, 34 pages, 7 annexes.

Vladykov, V.D. & P.K.L Liew 1982 Sex of adult American eels (Anguilla rostrata ) collected as elvers in two different streams along the eastern shore of Canada, and raised in the same freshwater pond in Ontario. pp.88-93. In: K.H. Loftus (ed) Proceedings of the 1980 North American Eel Conference, Ontario Ministry of Natural Resources, Ontario Fisheries Technical Report 4, Toronto.

Vøllstad, L.A., Jonsson, B. 1986 Life-History characteristics of the European eel Anguilla anquilla, in the Imsa River, Norway. Transactions of the American Fisheries Society 115, 864-871

Westin, L 1990 Orientation mechanisms in migrating European silver eel (Anguilla anguilla): temperature and olfaction. Marine Biology 106: 175-179.

Wirth, T. and Bernatchez, L 2001 Genetic evidence against panmixia in the European eel. Nature 409, 1037-1040

53 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Table 2.1.1 Recruitment data series. Part 1. Scandinavia and British Isles N S S S S DK D N.Irl. Irl Irl UK Imsa Göta Älv Viskan Motala Dalälven Vidaa Ems Bann Erne Shannon Severn 1950 2947 305 875 1951 1744 2713 210 719 1952 3662 1544 324 1516 1953 5071 2698 242 3275 1954 1031 1030 509 5369 1955 2732 1871 550 4795 167.00 1956 1622 429 215 4194 1957 1915 826 162 1829 1958 1675 172 337 2263 1959 1745 1837 613 4654 244.00 1960 1605 799 289 6215 7409 1229 1961 269 706 303 2995 4939 625 1962 873 870 289 4430 6740 2469 1963 1469 581 445 5746 9077 426 1964 622 181.6 158 5054 3137 208 1965 746 500 276 1363 3801 932 1966 1232 1423 158 1840 6183 1394 1967 493 283 332 1071 1899 345 1968 849 184 266 2760 2525 1512 1969 1595 135 34 1687 422 600 1970 1046 2 150 683 3992 60 1971 842 12 1 242 787 1684 4157 540 1972 810 88 51 88 780 3894 2905 1973 1179 177 46 160 641 289 2524 1974 631 13 58.5 50 464 4129 5859 794 1975 42945 1230 99 224 149 888 1031 4637 392 1976 48615 798 500 24 44 828 4205 2920 394 1977 28518 256 850 353 176 91 2172 6443 131 1.02 1978 12181 873 533 266 34 335 2024 5034 320 1.37 1979 2457 190 505 112 34 220 2774 2089 488 6.69 40.1 1980 34776 906 72 7 71 220 3195 2486 1352 4.50 32.8 1981 15477 40 513 31 7 226 962 3023 2346 2.15 32.0 1982 45750 882 380 22 1 490 674 3854 4385 3.16 30.4 1983 14500 113 308 12 56 662 92 242 728 0.60 6.2 1984 6640 325 21 48 34 123 352 1534 1121 0.50 29.0 1985 3412 77 200 15.2 70 13 260 557 394 1.09 18.6 1986 5145 143 151 26 28 123 89 1848 684 0.95 15.5 1987 3434 168 146 201 74 341 8 1683 2322 1.61 17.7 1988 17500 475 92 170 69 141 67 2647 3033 0.15 23.1 1989 10000 598 32 35.2 9 13 1568 1718 0.03 13.5 1990 32500 149 42 21 5 99 2293 2152 0.47 16.0 1991 6250 264 1 2 52 677 482 0.09 7.8 1992 4450 404 70 108 10 6 978 1371 0.03 17.7 1993 8625 64 43 89 7 20 1525 1785 0.02 20.9 1994 525 377 76 650 72 52 1249 4400 0.29 22.3 1995 1950 6 32 8 40 1403 2400 0.40 36.0 1996 1000 277 1 14 18 20 2667 1000 0.33 25.7 1997 5500 180 8 8 8 5 2533 1038 2.12 16.9 1998 1750 5 6 15 4 1283 782 0.28 20.0 1999 3750 2 85 16 3 1345 1100 0.02 18.0 2000 1625 14 270 12 4 563 900 0.04 7.6 2001 1875 2 178 8 1 250 699 0.003 5.4 2002 685 13 1000 112 0.16

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 54

Table 2.1.1 Recruitment data series; continued. Part 2: Mainland Europe. NL B F F F F F E P/E It Gironde Gironde DenOever Ijzer Vilaine Loire (CPUE) (Yield) Adour Nalon Minho Tiber 1950 6.56 86 1951 12.94 166 1952 83.88 121 14529 1953 12.22 91 8318 1954 18.32 86 13576 1955 25.15 181 16649 1956 6.68 187 14351 1957 14.98 168 12911 1958 47.75 230 13071 1959 26.73 174 17975 1960 20.67 411 13060 1961 35.7 334 17177 1962 80.48 185 11507 1963 115.38 116 16139 1964 36.42 3.7 142 20364 1965 75.26 115.0 5.0 134 11974 1966 17.9 385.0 4.0 253 12977 1967 28 575.0 9.0 258 20556 1968 19.06 553.5 12.0 712 15628 1969 16.16 445.0 10.0 225 18753 1970 36.49 795.0 8.0 453 17032 1971 16.52 399.0 44.0 330 11219 1972 28.99 556.5 38.0 311 11056 1973 22.26 356.0 78.0 292 24481 1974 24.53 946.0 107.0 557 32611 1.642 1975 32.14 264.0 44.0 497 55514 10.578 11.00 1976 25.51 618.0 106.0 770 37661 20.048 6.70 1977 56.8 450.0 52.0 677 59918 36.637 5.90 1978 37.47 388.0 106.0 526 37468 24.334 3.60 1979 50.32 675.0 209.0 642 19.7 286.2 42110 28.435 8.40 1980 25.79 358.0 95.0 525.5 25.9 404.8 34645 21.32 8.20 1981 21.59 74.0 57.0 302.7 20.0 332.2 26295 54.208 4.00 1982 13.71 138.0 98.0 274 15.0 123.3 21837 16.437 4.00 1983 8.98 10.0 69.0 259.5 13.6 80.3 22541 30.447 4.00 1984 12.3 6.0 36.0 182.5 19.2 82.0 12839 31.387 1.80 1985 13.79 13.0 41.0 154 9.6 64.5 13544 20.746 2.50 1986 14.42 26.0 52.6 123.4 10.6 45.2 8 23536 12.553 0.20 1987 5.71 33.0 41.2 145 14.0 82.4 9.5 15211 8.219 7.40 1988 3.88 48.0 46.6 176.6 10.9 33.0 12 13574 8.001 10.50 1989 2.71 30.0 36.7 87.1 7.2 80.0 9 9216 9.000 5.50 1990 3.29 218.2 35.9 96 5.6 48.1 3.2 7117 6.000 4.40 1991 0.99 13.0 15.4 35.7 7.7 64.0 1.5 10259 9.000 0.80 1992 2.6 18.9 29.6 39.3 3.7 41.7 8 9673 10.000 0.60 1993 2.61 11.8 31.0 90.5 8.2 69.4 5.5 9898 7.600 0.50 1994 4.48 17.5 24.0 94.6 8.7 45.8 3 12602 4.700 0.50 1995 6.52 1.5 29.7 132.5 8.2 73.2 7.5 5992 15.200 0.30 1996 7.38 4.5 22.4 80.8 4.8 30.7 4.1 3655 8.700 0.10 1997 11.9 9.8 22.6 70.8 6.5 50.5 4.6 3275 7.400 0.10 1998 2.12 2.3 17.5 60.7 4.3 25.0 1.5 3814 7.400 0.13 1999 3.27 15.3 86.9 7.5 44.1 4.3 1331 0.06 2000 1.62 17.9 14.2 79.9 10 1289 0.07 2001 0.53 1.0 8.1 0.04 2002 1.09 1.4 16.0 0.02

55 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Table 2.2.1 Statistics of eel landings, reported in the FAO database of fishing yields. These data include landings of ‘river eels’ in Atlantic waters, the Mediterranean and Inland waters. Data for several countries have been corrected, e.g. for erroneous inclusion of aquaculture. Country Norway Sweden Denmark Germany Ireland UK Netherlands France Spain Portugal Italy Remaining Northern year Europe Africa 1950 300 2188 4500 400 4200 500 100 1000 1951 300 1929 4400 400 3700 500 100 1000 1952 200 1598 3900 400 4000 700 100 1000 1953 400 2378 4300 500 400 3100 600 100 1000 900 1954 300 2106 3800 300 500 2100 500 900 1000 800 1955 500 2651 4800 500 700 1700 500 600 1000 1000 1956 300 1533 3700 400 600 1800 500 800 2000 900 1957 400 2225 3600 400 600 2500 500 500 2000 800 1958 400 1751 3300 400 100 600 2700 500 500 2100 1200 1959 400 2789 4000 500 100 500 3400 900 500 3000 700 1960 400 1646 4723 400 0 800 3000 1300 500 2700 1000 1961 500 2066 3875 500 100 800 2660 1300 400 2600 900 300 1962 400 1908 3907 400 100 700 1543 1300 800 3100 1000 300 1963 500 2071 3928 2100 100 700 1818 1400 1100 3500 1000 300 1964 400 2288 3282 1900 100 600 2368 1400 1700 3500 1100 400 1965 500 1802 3197 1500 200 800 2509 1700 1300 3200 900 500 1966 500 1969 3690 1700 100 1000 2739 1300 1300 3100 1000 400 1967 500 1617 3436 1900 100 600 2884 2000 1400 3100 1100 400 1968 600 1808 4218 1800 100 600 2622 2700 1300 3200 1100 400 1969 500 1675 3624 1600 100 600 2741 1900 1400 3400 1100 400 1970 400 1309 3309 1600 200 800 1512 4200 1100 3300 1400 100 1971 400 1391 3195 1300 100 800 1153 4900 1100 3400 1500 100 1972 400 1204 3229 1300 100 700 1057 2600 1000 2900 1138 100 1973 400 1212 3455 1300 100 800 1023 3900 700 2900 1150 800 1974 383 1034 2814 1285 67 817 994 2493 1300 42 2697 1528 352 1975 411 1399 3225 1398 79 833 1173 1590 570 44 2973 1400 85 1976 386 935 2876 1322 150 694 1306 2959 675 38 2677 1254 47 1977 352 989 2323 1317 108 742 929 1538 666 52 2462 1384 159 1978 347 1076 2335 1162 76 877 862 2455 655 44 2237 1357 112 1979 374 956 1826 1164 110 879 687 3144 394 25 2422 1518 134 1980 387 1112 2141 1051 75 1053 828 4503 300 32 2264 1242 448 1981 369 887 2087 1033 94 858 876 1425 250 33 2340 1192 497 1982 385 1161 2378 1027 144 1032 1097 1469 200 14 2087 1419 455 1983 324 1173 2003 1029 117 1113 1230 1856 150 11 2076 1782 575 1984 309 1073 1745 911 88 957 681 2336 150 80 2361 2445 477 1985 352 1140 1519 866 87 781 666 2288 200 76 1907 2123 258 1986 271 943 1552 887 87 997 729 2924 200 633 1928 1867 356 1987 282 897 1189 731 221 939 512 2378 259 566 2076 2479 306 1988 513 1162 1759 746 215 715 590 2879 205 501 2165 2790 256 1989 312 952 1582 678 400 1075 645 2482 83 6 1301 2365 368 1990 336 942 1568 976 256 1039 657 2484 75 295 1199 2209 560 1991 323 1084 1366 1010 245 822 707 2260 65 314 1106 2337 358 1992 373 1180 1342 1026 234 782 621 1964 60 674 1662 2749 358 1993 340 1210 1023 1027 260 752 320 1674 55 505 1307 2509 613 1994 472 1553 1140 585 300 873 369 1417 50 979 986 2797 732 1995 454 1205 840 585 400 808 279 500 106 10 886 2572 1176 1996 352 1134 717.5 696 550 895 336 563 97 21 883 2676 984 1997 497 1382 757.6 746 550 807 315 1942 113 16 1010 2034 1327 1998 353 645 557 717 670 741 346 491 160 13 682 2159 1069 1999 475 734 686 747 675 697 372 189 166 3 1532 1257 2000 281 561 620 686 250 796 368 247 48 29 604 30 2001 429 110 795 351

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 56

Table 2.3.1.1 Re-stocking of glass eel. Numbers of glass eels (in millions) re-stocked in (eastern) Germany (D east), the Netherlands (NL), Sweden (S), Poland (PO), Northern Ireland (N.Irl.) and Belgium (Flanders). D east NL S PO N.Irl. Flanders 1945 17.0 1946 7.3 21.0 1947 7.6 1948 1.9 1949 10.5 1950 0.0 5.1 1951 0.0 10.2 0.0 1952 0.0 16.9 0.1 17.6 1953 2.2 21.9 0.0 25.5 1954 0.0 10.5 26.6 1955 10.2 16.5 30.8 0.5 1956 4.8 23.1 21.0 1957 1.1 19.0 24.7 1958 5.7 16.9 35.0 1959 10.7 20.1 52.5 0.7 1960 13.7 21.1 64.4 25.9 1961 7.6 21.0 65.1 16.7 1962 14.1 19.8 61.6 27.6 1963 20.4 23.2 41.7 28.5 1964 11.7 20.0 0.0 39.2 10.0 1965 27.8 22.5 39.8 14.2 1966 21.9 8.9 69.0 22.7 1967 22.8 6.9 74.2 6.7 1968 25.2 17.0 12.1 1969 19.2 2.7 3.1 1970 27.5 19.0 12.2 1971 24.3 17.0 14.1 1972 31.5 16.1 8.7 1973 19.1 13.6 7.6 1974 23.7 24.4 20.0 1975 18.6 14.4 15.1 1976 31.5 18 9.9 1977 38.4 25.8 19.7 1978 39.0 27.7 16.1 1979 39.0 30.6 0.1 7.7 1980 39.7 24.8 0.1 11.5 1981 26.1 22.3 16.1 1982 30.6 17.2 24.7 1983 25.2 14.1 2.9 1984 31.5 16.6 12.0 1985 6.0 11.8 0.8 13.8 1986 23.8 10.5 0.1 25.4 1987 26.3 7.9 0.0 25.8 1988 26.6 8.4 0.2 23.4 1989 14.3 6.8 0.0 9.9 1990 10.65 6.1 0.7 13.3 1991 2.01 1.9 0.3 3.5 1992 6.36 3.5 0.3 9.4 1993 7.62 3.8 0.6 9.9 0.8 1994 7.6 6.2 1.7 16.4 0.5 1995 0.99 4.8 1.5 13.5 0.5 1996 0.05 1.8 2.3 11.1 0.5 1997 0.38 2.3 2.4 10.9 0.4 1998 0.3 2.5 2.1 6.2 0.0 1999 0.0 2.9 2.2 12.0 0.8 2000 0.0 2.8 1.2 5.4 0.0 2001 0.9 0.7 2.8 0.2 2002 1.6

57 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Table 2.3.1.2 Re-stocking of young yellow (bootlace) eel. Numbers of young yellow eels (in millions) re- stocked in (eastern) Germany (D east), the Netherlands (NL), Sweden (S), Denmark (DK) and Belgium (Flanders). D east NL S DK Flanders 1945 1946 1947 1.6 1948 2.0 1949 1.4 0.0 1950 0.9 1.6 0.0 1951 0.9 1.3 0.0 1952 0.6 1.2 0.0 1953 1.5 0.8 0.0 1954 1.1 0.7 0.0 1955 1.2 0.9 0.0 1956 1.3 0.7 0.0 1957 1.3 0.8 0.0 1958 1.9 0.8 0.0 1959 1.9 0.7 0.0 1960 0.8 0.4 0.0 1961 1.8 0.6 0.0 1962 0.8 0.4 0.0 1963 0.7 0.1 0.0 1964 0.8 0.3 0.1 1965 1.0 0.5 0.1 1966 1.3 1.1 0.1 1967 0.9 1.2 0.1 1968 1.4 1.0 0.1 1969 1.4 0.0 0.0 1970 0.7 0.2 0.0 1971 0.6 0.3 0.0 1972 1.9 0.4 0.1 1973 2.7 0.5 0.1 1974 2.4 0.5 0.1 1975 2.9 0.5 0.1 1976 2.4 0.5 0.1 1977 2.7 0.6 0.0 1978 3.3 0.8 0.1 1979 1.5 0.8 0.1 1980 1.0 1.0 0.1 1981 2.7 0.7 0.1 1982 2.3 0.7 0.4 1983 2.3 0.7 1.0 1984 1.7 0.7 0.8 1985 1.1 0.8 0.9 1986 0.0 0.7 0.5 1987 0.0 0.4 1.0 1.6 1988 0.0 0.3 1.3 0.8 1989 0.0 0.1 1.0 0.4 1990 0.1 0.0 1.6 3.5 1991 0.1 0.0 1.8 3.1 1992 0.1 0.0 2.2 3.9 1993 0.2 0.2 2.0 4.0 0.2 1994 0.2 0.0 2.0 7.4 0.1 1995 0.7 0.0 1.8 8.4 0.1 1996 0.9 0.2 2.5 4.6 0.1 1997 1.5 0.4 2.5 2.5 0.1 1998 1.2 0.6 2.4 3.0 0.1 1999 1.1 1.2 2.4 4.1 0.1 2000 1.0 1.0 1.5 3.8 0.0 2001 0.4 1.7 0.0 2002 0.4

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 58

Table 2.4.1 Production of European eel in aquaculture in Europe and Japan. Compilation of production estimates (tonnes) derived from reports of previous meetings, FAO, FEAP and others. Data for Denmark has been revised for 1984-2001.

1984 1985 1986 1987 1988 1989 1990 1991 1992 Norway Sweden 15 47 59 193 233 190 160 195 192 Denmark 18 40 200 240 195 430 586 866 748 Germany Ireland UK 20 30 0 0 Netherlands 20 100 200 200 350 550 520 1250 Belgium/Lux. 30 30 125 125 125 Spain 15 20 25 37 32 57 98 105 175 Portugal 60 60 590 566 501 6 270 622 505 Marocco 35 41 68 Algeria 72 53 22 1 0 Tunisia 150 151 250 Italy 2600 2800 4200 4600 4250 4500 3700 4185 3265 Greece 6 4 10 54 94 132 337 Turkey Macedonia 1 Yugoslavia 44 52 48 49 19 10 5 1 8 Croatia 7 5 Hungary 90 39 73 33 Czech.rep. 2 Sum EU 1950 2229 3448 4729 5517 5159 6667 6098 6818 Japan 3000

1993 1994 1995 1996 1997 1998 1999 2000 2001 Norway 120 200 200 200 200 Sweden 182 158 184 215 250 250 250 260 253 Denmark 782 1034 1324 1568 1913 2483 2718 2674 2040 Germany 100 100 100 150 150 150 150 300 160 Ireland 100 UK 25 25 Netherlands 1487 1535 2800 2443 3250 3800 4000 3800 3228 Belgium/Lux. 125 150 140 150 150 40 20 50 55 Spain 134 214 249 266 270 300 425 200 Portugal 979 200 110 200 200 200 200 Marocco 85 55 55 56 Algeria 22 20 17 17 Tunisia 260 108 158 147 108 Italy 3000 2800 3000 3000 3100 3100 3100 2750 2500 Greece 341 659 550 312 500 500 300 600 Turkey Macedonia 0 70 83 60 Yugoslavia 2 9 5 5 Croatia 5 7 6 7 Hungary 50 50 19 19 Czech.rep. 4 4 3 3 Sum EU 7721 7689 8935 9031 10646 11059 10839 10510 8435 Japan 10000

59 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Table 5.3.2 Actual damage rates of eels estimated at hydropower stations Author(s) % All sizes of eels: BERG (1985) 36.7 BERG (1988) 9.31 *) BERG (1993) 15.4-25 BERG (1994) 30.4-40.5 KISKER (1930) 12.5 LUNDBECK (1927) 5.5 Von RABEN (1955) 18.4-19.6 BUTSCHEK & HOFBAUER (1956) 12-40.5 WONDRAK (1989) 54-87 SEIFERT (1989) 42-50 DESROCHERS (1995): (cit. in McCLEAVE 2001) 16 - 24 HADDERINGH (1996) 5-25 HADDERINGH & BAKKER (1998) 13.5 HOLZNER (1999) 27 DÖNNI, MAIER & VICENTINI (2001) 17 - 86 LINDROTH (1941); SVÄRDSON (1944) (cit. in LARINIER & DARTIGUELONGUE (1989) 40 Mean 28.5 Silver eels : GUSTAVSBERG & MAI (1960) (cit. in LARINIER & DARTIGUELONGUE (1989) 91 – 100 1) LANGGÖL (1960, 1961) (cit. in LARINIER & DARTIGUELONGUE (1989) 75 – 80.8 1) LARINIER & DARTIGUELONGUE (1989) 40 – 63 1) LARINIER & DARTIGUELONGUE (1989) 51 – 92 1) LARINIER & DARTIGUELONGUE (1989) 81.2 1) LARINIER & DARTIGUELONGUE (1989) 63 2) LARINIER & DARTIGUELONGUE (1989) 100 3) MONTEN (1985) (cit. in McCLEAVE 2001) 40 – 100 1) RMC (1995) (cit. in McCLEAVE 2001) 9 1) 3) Mean 68.8 1) silver eels 73 - 90 cm; 2) silver eels 56.5 cm ; 3) Francis turbines; *) considerable reduced water flow

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 60

Table 5.3.7 European electricity generation (source : www.iea.org/public/allpubs.htm)

Electricity generation in IEA Countries Generation 1998 Capacity Hydro Total Hydro Hydro 1998 2005 2010 TWh % TWh GW net GW net GW net Austria 55,9 66,5 37,2 11,44 11,66 11,66 Belgium 82,1 0,5 0,4 1,40 1,40 1,40 Czech Republic 64,6 2,2 1,4 2,03 .. .. Denmark 41,1 0,1 0,0 0,01 - - Finland 70,2 21,4 15,0 2,88 2,89 2,89 France 506,9 12,2 61,8 25,10 25,10 25,10 Germany 552,4 3,1 17,1 8,85 10,44 10,64 Greece 46,2 8,0 3,7 2,86 3,46 3,73 Hungary 37,2 0,4 0,1 0,05 0,05 0,05 Ireland 20,9 4,4 0,9 0,53 0,54 0,54 Italy 253,6 16,3 41,3 20,06 20,78 21,30 Luxembourg 0,4 31,2 0,1 1,14 1,14 1,14 Netherlands 91,2 0,1 0,1 0,04 0,05 0,05 Norway 116,1 99,4 115,4 28,02 .. .. Portugal 38,9 33,4 13,0 4,50 4,81 5,01 Spain 193,5 17,6 34,1 16,63 .. .. Sweden 158,2 47,0 74,4 16,26 16,30 16,70 Switzerland 61,7 54,2 33,4 11,98 .. .. Turkey 111,0 38,0 42,2 10,31 15,45 21,47 United Kingdom 356,6 1,5 5,3 4,26 4,30 4,30 IEA Europe 2858,7 17,4 497,4 168,35 .. ..

Table 7.4 Data on the relative proportion of eel in the food of cormorants.

Reference Eel as percentage Comments of food Hald-Mortensen 1995 2 Denmark 1992-94 Adam 1997 19 France Knösche 2002 36 Germany summer Knösche 2002 2 Germany winter Madsen and Spärck 1950 22 Denmark 1942-48

Table 7.6 Summary of management alternative for increase of SSB of eel.

Management alternative Scoop for increase of Cost re. fishery SSB regulation (ton/y) Fishery regulation 14 000 n.a. Safe fish passage up 2 000 Low Safe fish passage down 2 500 – 10 000 Medium-high Predation birds 1 800 – 9 000 Low Predation mammals 4 000 Not acceptable Stocking 60 000 Low

61 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

APPENDIX 1. LIST OF PARTICIPANTS.

Willem Dekker (chairman) Thomas Changeux RIVO Conseil Supérieur de la Pêche - Drection Générale Haringkade 1 134, avenue de Malakoff PO Box 68 75116 Paris 1970 AB IJmuiden France Holland phone: +33145022005, fax: +33145012723 phone: +31 255 564 712, fax: +31 255 564 644 e-mail: [email protected] e-mail: [email protected] Paula Cullen Miran Aprahamian Zoology Department Environment Agency National University of Ireland, Galway Northwest Region Galway Richard Fairclough House Ireland Knutsford Rd phone: +353 (0) 91 524411 ext 3360 , fax: +353 (0) 91 UK 750526 phone: +44 1925 653 999, fax: +44 1925 415 961 e-mail: [email protected] e-mail: miran.aprahamian@environment- agency.gov.uk Eric Feunteun University of Rennes I, UMR CNRS 1853, Claude Belpaire Laboratoire d'Evolution des Systèmes Naturels & Institute for Forestry and Game Management Modifiés Duboislaan 14 Avenue du Général LECLERC 1560 Groenendaal Hoeilaart 35042 Rennes Cedex Belgium France phone: +32 2 6570386, fax: +32 2 6579682 phone: +33 299 28 14 39, fax: +33 299 28 14 58 e-mail: [email protected] e-mail: [email protected]

Araitz Bilbao Batiz Jan Klein Breteler AZTI - Technological Institute for Fisheries and Food, OVB Department of Fisheries Resources, Postbus 433 Txatxarramendi Ugartea z/g 3430 AK Nieuwegein 48395 Sukarrieta Holland Spain phone: +31 30 605 8445, fax: +31 30 603 9874 phone: +34 946029400, fax: +34 946870006 e-mail: [email protected] e-mail: [email protected] Reiner Knoesche Cedric Briand Institut fur Binnenfischerei Institut dAménagement de la Vilaine Potsdam - Sacrow 56 130 La Roche Bernard D 14476 Gross Glienicke France Jaegerhof phone: +33 2 99 90 88 44, fax: +33 2 99 90 88 49 Germany e-mail: [email protected] phone: +49 33 201 40630, fax: +49 33 201 40640 e-mail: [email protected] Gerard Castelnaud Inland Living Aquatic Resources Research Unit Najih Lazar Cemagref, Agricultural and Environmental RI Div of ish and Wildlife Engineering Marine Fisheries Reseach Regional Centre of Bordeaux 3 Fort Wetherill Road 50, av. de Verdun Jamestown France USA phone: +33 5 57 890 803, fax: +33 5 57 890 801 phone: +1 401-4236-1926, fax: +1 401-423-1925 e-mail: [email protected] e-mail: [email protected]

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 62

Laura M. Lee Eric Rochard Atlantic States Marine Fisheries Commission Cemagref c/o Rhode Island Division of Fish and Wildlife Unité Ressources Aquatiques continentales Marine Fisheries 33612 CESTAS Cedex 3 Fort Wetherill Road France Jamestown phone: +33 5 57 89 08 13, fax: +33 5 57 89 08 01 USA e-mail: [email protected] phone: +1 401 423 1935, fax: +1 401 423 1925 e-mail: [email protected] Robert Rosell DARDNI Ken Oliveira Newforge Lane University of Massachusetts Dartmouth Belfast BT9 5PX Department of Biology BT9 5PX 285 Old Westport Road UK North Dartmouth MA, 02747 phone: +44 2890 255 506, fax: +44 2890 255 004 USA e-mail: [email protected] phone: +1 508-999-8227, fax: +1 508 999 8196 e-mail: [email protected] Guy Verrault Société de la Faune et des Parcs du Québec Mike Pawson 506, rue Lafontaine CEFAS Rivière-du-Loup. Qc. Pakefield Rd G5R 3C4 Lowestoft NR33 0HT Canada Suffolk phone: +1 418.862.8649 #226, fax: +1 418.862.8176 UK e-mail: [email protected] phone: +44 1502 524436, fax: +44 1502 524511 e-mail: [email protected] Håkan Westerberg Nat. Bd. Fisheries Michael Pedersen PO Box 423 DFU SE 401 26 Göteborg Afd. for Ferskvandsfiskeri SE-40126 Göteborg Veilsovej 39 Sweden 8600 Silkeborg phone: +46 31 743 0333, fax: +46 31 743 0444 Denmark e-mail: [email protected] phone: +45 33 96 31 00, fax: +45 33 96 31 50 e-mail: [email protected] Håkan Wickström Institute of Freshwater Research Russell Poole SE - 178 93 Drottningholm Marine Institute Sweden Furnace phone: +46 86200407, fax: +46 875 90338 Newport e-mail: [email protected] Co. Mayo Ireland Beth Williams phone: +353 98 42300, fax: +353 98 42340 Division of Life Sciences e-mail: [email protected] Kings College London 150 Stamford Street Patrick Prouzet Waterloo Institut Français de Recherche pour l’Exploration de la UK Mer (IFREMER) phone: +44 171 848 4108, fax: +44 848 4500 Laboratoire Halieutique d’Aquitaine (LHA) e-mail: [email protected] Technopole IZARBEL 64210 Bidart France phone: +33 5 59 41 53 96, fax: +33 5 59 41 53 59 e-mail: [email protected]

63 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

APPENDIX 2: OVERVIEW OF HABITAT CHANGES PER COUNTRY.

1 INTRODUCTION

The information in this section is obtained from www.small-hydro.com (data on hydropower production), Moriarty and Dekker (1997)(eel and topographic data) and Dill (1993)(topographic and hydropower data).

2. NORWAY

2.1 Hydrology and Topography

Most of the rivers in Norway have small catchment areas (rarely more than 3000 km2), are short and swift, and drain westerly into the Atlantic. Western streams generally run precipitously from the mountains to the valley and then rapidly to their mouths. The many waterfalls block the ascent of anadromous fish. Only in the south-east and north are the rivers of any length. Stream flows are generally high with respect to the area drained. For 38 rivers, each draining more than 1000 km2 and representing in all more than half the country, the average total discharge is no less than 4000 m3/s. Variations in flow are enormous, ranging sometimes between 1 and 3000 percent of the average.

2.2 Eel Habitat

No information is available from publications or reports. However, total landings from Norway amount to 475 tons of eels. Most of these eels probably grow up in the large areas of marine waters. Due to the steep slopes of most rivers and the large highland areas (fjells) the relevance of inland eel habitat is small.

2.3 Hydropower

In 1985 there were 625 hydroelectric power stations in Norway. More than 99 percent of electricity in Norway comes from hydropower, mostly high-head. Representing the fifth largest hydropower system in the world and regarded by many as the most efficient and modern, the Norwegian hydropower system generates an average of 11 500 GWh of electricity a year and has a total installed capacity of 29 932 MW.

2.4 Conclusion

Although hydropower is extremely highly developed in Norway, the impact on the local eel stock probably is not high, as most of it is produced in marine waters and most of the inland waters are difficult or not accessible naturally for eels.

3. SWEDEN

3.1 Topography and Hydrology

The total area is 487 000 km2 and thus Sweden is one of the largest countries in Europe. Sweden has more than 100 000 lakes covering roughly 10% of its surface. There are many rivers, of which thirteen have a mean annual flow of more than 100 m3/s at the mouth.

The highest precipitation is in the mountain regions where it can reach the figure of 2,500 mm per year in some places. Precipitation is also quite high in the south-western part of the country - somewhat more than 1,000 mm per year. The interior of Northern Sweden and the lowland areas in the middle and in the South are quite dry and have less than 400 mm precipitation per year. Evaporation is greatest in the south of the country, roughly 400 mm per year, and decreases northwards and as the altitude becomes higher. In the mountain areas it is only 100-200 mm per year.

The farther northwards one goes and the higher the altitude is, the greater part of the precipitation fails as snow. From the point of view of hydropower, it is of course an advantage that the precipitation is high and the evaporation is low in the mountain areas and in the highlands. On the other hand it is a disadvantage that the precipitation in these areas remains on the ground in the form of snow and ice for a lengthy period of the year and then flows away in a short time and with a high intensity during the spring.

The differences in altitudes are not very great in Sweden. In the northern part the rivers descend slowly from the mountain range in the west to the Gulf of Bothnia and the Bothnian Sea. In the mountain area the landscape slopes

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 64

steeply. Before the mountain streams have been united into small rivers the ground has descended to an altitude of 600- 800 m. On their way to the sea the rivers here and there form long narrow lakes.

In Sweden only three major rivers longer than 150 km and six minor rivers have not been affected by dams (www.iea.org).

3.2 Eel habitat

There are about 39,639 km2 of freshwater lakes in Sweden. Of this about 17,570 km2 seem suitable for eel production. It is estimated that about 11,074 km2 today support eel stocks which, however, in most cases are extremely sparse. There are large parts of this area which do not support any local eel fishery at all, due to very sparse stocks, but on the national scale most of them will contribute to the commercial fishery in the large lakes and along the Baltic Coast. A more developed eel fishery takes place in about 4,921.4 km2.

Coastal water. Eel are present in most Swedish coastal waters, ranging from quite abundant along the west coast (8,600 km2 with depths less than 20 m) to more and more scarce in the northern part of the Baltic Sea (25,500 km2 with depths less than 10 m). The total area is thereby 34,100 km2. There are commercial fisheries for eel up to about 60° N on the Baltic coast, so that eel fishing takes place in about 14,600 km2 out of the available 25,500 km2 along the Baltic coast.

Rivers/canals. There are 118 main rivers in Sweden and some additional 5,763 rivers and streams of sufficient size to be registered with their own co-ordinates in the database of the Swedish Meteorological and Hydrological Institute. Total length of all 5,881 is about 79,153 km. Assuming a mean width of 10 m, this length represents an area of about 792 km2. There are no data on how much of this is utilised by eel. However, there are no fisheries targeting eel in running water, except in small mill traps and some few fixed installations of the stow net type ållana.

Lakes. In rivers where ascending eel are still fairly abundant there are normally some kind of eel pass or upstream transportation. In that way most areas are theoretically accessible for eel. However, in areas where recruiting eel have become naturally rather few in recent days such installations have often been forgotten and mismanaged leaving upstream areas without eel. The extent of this is not known. Other possible obstacles to eel recruitment are low temperatures, low productivity and high altitudes in the northern parts of Sweden. Some parts of the largest lakes are too deep (>20 m) for eel. Considering this criterion we estimate that 20,620 km2 of our total lake area, i.e. 52%, have no potential for eel production. Today physical obstacles, etc. are perhaps not the primary limitation of eel in Sweden. The general lack of recruits, especially in the Baltic part of the country, is probably much more important.

Rivers/Canals The same observations apply as in the preceding paragraph.

3.3 History of Hydropower

The first generating stations based on hydropower were established in the 1880's. These stations were usually built where there had previously been directly driven machinery for mills, saws, hammers etc. These stations were small and were essentially intended to supply power to industries and communities in the immediate vicinity. Hundreds of such small local hydroelectric power stations were constructed during the end of the nineteenth and the beginning of the twentieth century.

As the technique of transferring power over longer distances developed in the beginning of the 1900's, it became possible to exploit the large rivers in the south and in the middle of the country. Until 1967 the power supply was almost entirely based on hydropower. In 1995 hydropower accounted for 47% of the production. The installed generating capacity at the end of 1995 was 16,150 MW in hydroelectric stations.

3.4 Hydropower Developed Until 1995

The total annual hydropower production today amounts to somewhat less than 64 TWh. In all there are about 1,000 hydro stations in the system, fourteen of which have an installed capacity of more than 200 MW. About 650 stations are smaller than 1 MW and contribute together in the order of 1 TWh per year only.

The hydropower is highly regulated. Almost half of the hydropower produced comes from water stored for shorter or longer periods in the reservoirs. The production in mid winter is up to ten times that corresponding to the natural runoff of the rivers.

65 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

3.5 Available Hydropower Potential

On the basis of data relating to the topography and runoff, the total natural potential has been estimated at 200 TWh per year. Considering inevitable losses, efficiency of machinery and water-ways, unfeasible sites etc, it has been judged technically possible to develop about 130 TWh per year.

On several occasions throughout the years studies have been made as to what extent hydropower could be developed economically. Each new estimate has produced a higher figure than the previous one. This is caused partly by advances in design and construction and partly by changed costs of alternative power production. The latest study was done in 1995 and gave the figure 95 TWh per year as economically feasible. In the fifties and sixties the evaluation was made in comparison with oil-based production of power. The latest estimate is based on comparison with alternative renewable sources of electric energy. With an economically feasible potential of at least 95 TWh per year, of which 64 TWh is developed today, there should be an additional resource of 31 TWh. The bulk of these resources is located in northern Sweden. About 17 TWh per year relates to the four rivers Vindeldlven, Pite Alv, Kalix Alv and the Swedish section of the river Torne-Muonio Alv, which are mainly undeveloped.

3.6 Options for further development

From the forties until the mid-sixties, hydropower was developed in several steps with the increasing consumption of electricity. Thermal power stations were built exclusively in order to safeguard the power supply in dry years and to cope with the peak loads. The later fifties, however, saw a growing resistance to the development of hydro power. In the sixties interest in the conservation of the environment was broadened in the general debate and resulted for example, in the creation of the Environment Protection Board. As a result, the further development of hydropower was called into question in a very different way than previously.

With the hopes attached to nuclear power development, a new argument was advanced: hydropower might no longer be necessary. The additional power to be gained from an entire river appeared almost marginal compared with the energy production of large nuclear stations.

Views as to what extent further hydroelectric development should be permitted still differ widely. The main argument in favour of further development is that the utilization of hydropower is an economic and efficient method of generating electricity It normally does not cause any significant air or water pollution and is a domestic resource which should be utilized to contribute to reducing dependence on imported fuel, above all oil. Others, primarily the nature conservancy authorities and organizations, declare that no further development of hydro power should be considered because, in their opinion, exploitation of the remaining, untouched rivers threatens to deprive Sweden entirely of the type of natural country these represent.

The possibilities of developing hydropower in Sweden now depend not so much on technical and economic considerations, as on the degree to which new projects can be accepted in view of their effects on the environment. The consequence is that the Government, during the last decades, has made the final judgements by weighing the advantages and disadvantages of the projects. To obtain a basis for its decisions, the Government has commissioned extensive studies during the seventies, eighties and nineties designed to:

• take aside those of specific interest for harnessing. • list the possibilities of further, economical rational, hydro power developments. • describe the consequences of these for different environmental interests. • rank these projects in respect of both their environmental effects and their economic advantages in power production.

On the basis of these studies, the Government and the Riksdag have drawn up guidelines for the future utilization of Sweden's rivers for hydropower. These entail that the major part of those rivers, and individual stretches of rivers, which have not yet been claimed for development will also remain undeveloped. Consequently the four main rivers Vindelalven, Pite Alv, Kalix Alv and Torne-Muonio Alv amongst others, are excluded. What then remains to be developed, if permission by the Water Rights Court is granted, corresponds to an annual power production of less than 2 MW. The majority of these projects are among the least economically advantageous and include the so called “mini power stations”. It is therefore uncertain whether they will be developed in the foreseeable future.

It may well be doubted whether Sweden can afford in the long run to leave unexploited about one-third of the hydropower potential that is estimated to be economically feasible today and about half of the technically possible resources. Bearing in mind the political situation and the attitude of large groups of the people, several years will

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 66

certainly pass before a substantial further hydropower development programme is reconsidered. Thanks to good flow regulation conditions, it has been and is still possible to increase the capacity of many hydro stations and to use these for peak load production. In this way it is also possible to maintain a sufficient reserve capacity of hydropower to compensate, for instance, for the temporary shutdown of a large nuclear unit. There are only two pumped storage stations and no more are foreseen in the main system within the near future.

3.7 Conclusions

Sweden is a very large country with numerous lakes and rivers with useful and naturally accessible eel habitat. Most of the rivers are obstructed however by dams for hydroelectric energy and the turbines are affecting eel survival at downstream migration. Notwithstanding the fact that there are large coastal areas producing eel as well, the effect of dams and hydropower on the local eel stock probably is relatively high.

4. FINLAND

4.1 Eel habitat

Available eel habitat in Finland has not been described. But numerous rivers and lakes are naturally accessible for eels and the coastal areas may also contribute to the available eel habitat.

4.2 Government Regulation of Hydropower

Hydropower plants in Finland are not directly owned by the State. The largest state-owned (Or mainly state-owned) hydro companies with hydropower Ownership are Imatran Voirna Oy and Kernijoki Oy. The share of these two companies of hydropower production is 60 %. The next largest producers are hydropower plants mainly owned by the wood processing industry. The rest are mainly owned by communes of local electricity distribution companies.

Controlling of hydropower construction in Finland is realized in accordance with energy policy-related programmes of the Government. Finnish hydropower projects are not regulated in detail by the State. The hydropower company submits an application to the Water Rights Court, where the handling of permits and permit conditions of individual projects are stipulated. The permit conditions are then supervised by regional environment centres.

4.3 Energy Policy and Hydropower

The Finnish State neither has any special regulations or definitions for "small or large-scale-hydro" nor any programmes which slow down or discourage hydropower development. Some watercourses are, however, protected from hydropower construction by special Act (stipulated in the 1980s).

The present Government has not come up with any energy policy related report. The previous Government stated in a report published in 1992: "Hydropower will be increased on such already constructed watercourses where environmental viewpoints allow it. Adjustments in the regulations governing watercourses do not essentially decrease the amount of available hydropower potential. Construction of small hydro is being promoted. Decisions concerning reservoirs - which result in better utilization of hydropower - are made after the completion of reports on financial and environmental impacts." The alteration of taxation (passed by the Parliament in 1996) removed the liability tax on the production of all types of electricity generation, thus weakening the economic competitive position of renewable energy.

4.4 Industry Organization

In 1996, a Central Organization for the Energy Industry was founded in order to attend to the interests of the energy sector. It is yet too early to estimate the importance of the hydropower sector in this organization.

4.5 Electricity Production in Finland

Hydropower capacity has increased slowly, but its share of overall electricity production has decreased. The 1998 IEA- data show that 21.4% of its electricity generation is produced by hydropower stations, totalling 15.0 TWh.

67 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

4.6 Environmental Impact Assessment (EIA) Act

The Act concerning Environmental Impact Assessments was passed in Finland in 1994. According to the Act, the environmental impacts of such major construction projects which can have a considerable effect on the environment or cause health problems, will have to be assessed. The Act covers reservoir with an area of more than 10 km2. In addition, it concerns such regulation projects where the mean flow of the watercourse is more than 20 m'/s and the flow and water level conditions change radically. The EIA procedure should be applied for dams that can cause danger to people, their health or property. In Finland the Ministry of Forestry and Agriculture determines the classification of the dams.

4.7 Licensing Procedures

Construction of hydropower plants and reservoirs requires an official permit which is stipulated in the judicial procedure. The authorities are the Water Rights Court (WRC) (the first instance) and the Water Rights Appeal Court (the second instance). These are independent courts of law, and they are also independent of the administrative authorities.

The legal process of a hydropower project starts with an application submitted by the hydro company to the WRC. This Court arranges for the inspection proceedings to check the construction plan. The idea of the inspection proceedings is to allow those whose interests or rights are affected by the project to present their opinions and claims.

After the inspection proceedings it is up to the WRC to decide whether or not the applicant has a legal right to implement the project. According to the Law on Water Rights, if the project does not harm or damage the public or private interest, and if the public interest calls for construction, the permit must be granted. The issue related to sharing of water resources is also handled in this process.

4.8 The Vuotos Artificial Lake Project

The Vuotos artificial lake project has been under lively discussion for over 30 years now. In 1982, the Government ceased the project, but the Government of 1992 started it again. In 1995 the Government decided to proceed with the project. The Vuotos project is now being considered by the Water Rights Court which determines whether there are - according to the present legislation - grounds for construction and on which terms. When the decision has been given, the Government will take a stand on further measures.

4.9 Conclusion

There is a lack of data on natural eel production. Hydropower production is well developed in Finland. Possibly it is affecting the local eel stock seriously.

5. DENMARK

5.1 Topography and Hydrology

The expression semi-inland waters is used to include fjords, estuaries, sheltered bays and lagoons covering some 3,000 km2 which possibly have relatively large eel stocks compared to more open coastal waters (Moriarty & Dekker 1997).

There are about 500 freshwater lakes and ponds in Denmark covering a total area of about 430 km2 and about 10 km2 of reservoirs.

There are no long or important rivers in Denmark because of its small size, low elevation, relatively low rainfall and absence of discharge from upstream countries. There are about 15,000 km of major rivers and streams. The total area including minor streams is 150 km2.

5.2 Eel habitat

Coastal waters Eel fishing takes place all along the Danish coasts. The only known area with no eel fishing is on the west coast of Jutland from the north tip down to the Wadden Sea. In the Wadden Sea, eel populations are present as they are on practically all other coasts. The area of coastal waters where eel stocks may be present is in the order of 10,000 km2 of which 9% is considered to be part of the Baltic Sea.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 68

Rivers/streams Eel have access to all streams and rivers because fish passes must be installed at any obstruction. Streams contaminated with ochre make up some 5–10% of the total river system. Due to low invertebrate populations in these streams only small populations of eel are present. In the largest river system, Guden å, the lakes are known to act as sinks for ascending elver and yellow eel and the tributaries upstream of the large lakes are devoid of eel.

Lakes Eel are believed to be present in most of these waters, either naturally recruited or stocked by landowners or local fishermen’s organisations. Eel passes are enforced by law where there is any obstruction..

5.3 Hydropower

The production of hydropower in Denmark comes from 120 power stations and according to the 1998-IEA data (www.iea.org) totals only 0.0 TWh (0.1 % of total electricity generation).

5.4 Conclusions

Most of the eel habitat in Denmark is in its coastal area that is relatively large as compared to the land surface. Although there are quite a number of hydropower stations, their effect on the local stock probably is not very high.

6. GERMANY

Germany covers an area of 356 974 km2.

There were 311 large dams in operation in 1993.

6.1 Eel habitat

6.2 Extent of water currently supporting eel stocks

Coastal waters and estuaries An exact figure for the area is difficult to obtain. The estimate for the North Sea is 1,800 km2 and for the Baltic 900 km2 , including the area of estuaries. Data are available for the following biotopes: Elbe estuary, Elbe estuary Cuxhaven, Hamburg Harbour, Elbe weir Geesthacht, lagoon Wyk/Föhr, North Sea near Elbe estuary, estuary of river Oste, rock littoral of Helgoland.

For the Baltic Sea data are available for the stock in the Oder-estuary connected with the Oder-lagoon and from the open Baltic. In Germany during the best years (1950s/1960s), the total catch from the North Sea and the Baltic was much higher than from inland waters. Eel fishing along the whole region of the German coast, especially in the brackish waters of the Baltic coast is important (yellow and silver eel).

Coastal lagoons On the German Baltic coast there is only one coastal lagoon, the legendary Conventer See which many years ago had a maximum yield of 60 kg eel ha-1. In the 1970s land reclamation took place which cut the lagoon off from the Baltic causing a severe decline in eel catches. Natural immigration of eel is no longer observed.

Lakes, rivers, canals the total inland water area of Germany is 4,000 km2 from which 3,000 km2 are suitable for eel (75%). The official fisheries census (1993) states that the total area used by commercial fishery is 480 km2 for rivers and 1,820 km2 for lakes. This amounts to a total of 2,300 km2. The area used by recreational fishery is not known.

The eel is the main species for commercial fishery in lakes and rivers.

6.2.1 Areas of water inaccessible to eel due to chemical or physical obstructions

Lakes No details have been published. A rough estimate is that about 5% of all lakes are influenced by physical obstructions; no chemical pollution, as a barrier to eel ascent, is known.

Rivers/canals Most of the southern rivers have several dams. For example, there are 12 dams in the German part of the River Mosel with a catchment of 30,000 km2.

69 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

The eel, with its wide ecological adaptability, will find living conditions and food in the whole German water area, and also in polluted rivers and in reservoirs.

6.3 Energy and power sectors

Total electricity consumption was 500 095 GWh in 1996. Hydropower provided 3.9 % of the total electrical energy production. Germany imported 37 404 GWh of electricity in 1996 and exported 42 670 GWh.

6.4 Hydropower development

The gross theoretical hydropower potential of Germany is 120 000 GWh/year (evaluated in 1991). The technically feasible potential is 25,5 TWh/year ( 1998), and the economically feasible potential is about 20 000 GWh/year (1991 ). In 1998 about 73 percent of the technically feasible potential has been developed (Bunge et al. 2001).

The installed capacity of all powerplants in operation in Germany is 114 068 MW, of which 4304 MW is hydro capacity. Hydro plants generated 17 285 GWh in 1996. A further 9 MW of hydro capacity is under construction.

There is 4636 MW of capacity at pumped-storage plants, and a further 1060 MW under construction. Pumped-storage plants generated 4381 MW in 1996, and consumed 5892 GWh.

The 1060 MW Goldisthal pumped-storage plant in eastern Germany is being developed by Vereinigte Energiewerke AG (VEAG). Construction began in October 1997 and is scheduled to be completed during 2003.

6.5 Small hydro

There are some 6000 small hydro plants (of less than 10 MW) in operation, with a total capacity of approximately 1300 MW. About 100 MW of further small hydro capacity is envisaged.

6.6 Conclusion

Germany is a large country and most of its rivers and lakes were originally accessible for eels. Much rivers are dammed and provided with hydropower stations. Its effect on the local stock probably is relatively high.

7. UNITED KINGDOM

7.1 Eel habitat

7.1.1 Extent of water currently supporting eel stocks

Estuaries Estuaries form a substantial part of the coastal zone of Britain, with 155 larger estuaries comprising an intertidal/channel area of ca. 5,300 km2. Fyke nets, traps and pair trawls have been or are currently being used to exploit eel stocks in shallower, warmer, productive waters with depositing shores in the south and east, such as the Humber, Wash, Solent, Severn and Dee. The Thames is the only well-defined estuarine fishery for which information is available (Naismith and Knights 1993). However, the area able to support reasonable eel stocks is probably 40–50% of the total, i.e. ca. 2,500 km2.

Coastal lagoons There are no significant coastal lagoons in Great Britain.

Coastal waters Eel stocks are high enough to support occasional trawling in shallower coastal waters in the southern North Sea and English Channel. Eel also appear in by-catches in other near-coastal waters. Taken together with the estuaries above, a conservative estimate of the total area of saline waters occupied by eel would be 5,000 km2.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 70

Lakes

Area (km2) Number < 1 1,665 1–10 50 10–100 2

The total area of lakes which could support eel stocks is ca. 1,924 km2.

Rivers/canals The total area of inland water is 2,404 km2, of which still waters account for about 80%, thus the area of rivers and canals which could support eel is about 500 km2. Estimates of length of river vary from 38,820 to 58,380 km for 155–160 main rivers, 1,445 rivers in total.

7.1.2 Areas inaccessible to eel due to chemical or physical obstructions

All inland waters, even acid ones, should be capable of supporting eel if access is possible. Relatively few are totally inaccessible.

Lakes Man-made lakes number approximately 500. Access to some of these, and to some natural lakes, is restricted because of isolation from rivers or by water storage and river regulation purposes. The area totally inaccessible is, however, believed to be relatively small.

Rivers/canals There are few totally unpassable obstructions of any significance, mainly because of the lack of major hydro dams in comparison with other European countries. Large tidal barriers built in recent years have had to be fitted with fish passes. However, most main rivers are carefully regulated for flood control, water supply and navigation, with numerous weirs and sluices (plus some tidal-flap or gate barriers to the sea in low-lying areas such as East Anglia and the Somerset Levels). The number and relative severity of such obstructions have been shown to inhibit elver and yellow eel immigration and hence full utilisation of catchments. Data have been published for the Thames, Severn and some smaller rivers.

This work has shown that upriver sites may not be utilised or stocks may achieve lower densities and productivities than might be expected. For example, in surveys of the Severn, 93% of sites on tributaries feeding the estuary but only 69% of those above the tidal limit contained eel. Comparable figures for the Thames are 62% and 34%, respectively. Of 181 streams 65% with pH >6.1 contained eel. Thus 60% could be used as an approximation of the area of riverine waters actually inhabited by eel in Great Britain. Lower density populations deeper in catchments tend to be dominated by older, larger and female eel. These can make a potentially high contribution to the total eel breeding stock.

7.2 Hydropower

Most of the generation takes place in Scotland, which is both mountainous and wet, with Wales also making a contribution. England and Northern Ireland have many quite small hydroelectric projects but their total contribution is weak.

7.2.1 History of Hydropower development

The great period of hydropower construction occurred in the years after the Second World War - from 1948 to 1965, when most of the Scottish schemes were built by the North of Scotland Hydroelectric Board. Since that time there has been little further construction due to increasing opposition to reservoir construction, on environmental grounds. Today, there are over 50 main hydropower stations in Scotland with an installed capacity of around 1050 MW. In addition there are three pumped storage power stations with a combined capacity of 830 MW. In Wales the Dinorwig pumped storage scheme has an installed capacity of 1200 MW and there are two other main hydro stations providing 16 MW.

7.2.2 Hydropower Potential

Various estimates have been made about how much potential hydropower is available for development in the U.K. There is probably a further 750 MW of capacity in Scotland, but it is difficult to say how much of this would be economically viable and equally importantly, environmentally acceptable. So far as small hydro is concerned, there is a

71 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

potential for about 300 MW of plants less than 10 MW, but probably not more than one third of this would be economically exploitable.

7.2.3 Organization of the Electricity Sector

As part of the process of privatising the U.K's electricity industry, an Office of Electricity Regulation (OFFER) was established.. OFFER has powers to regulate prices to ensure they comply with a formula based on the Retail Price Index. It can require individual companies to sell generating capacity, if it judges that competition is being adversely affected by monopolistic tendencies. The whole of the hydroelectric generating capacity in the U.K. is now privately owned, including the pumped storage stations. The national grid has also been privatised, and in 1998 the electricity market is to become completely open to competition, though it will continue to be regulated by OFFER.

There is a programme for the subsidising of renewable energy (including small hydropower under 5 MW installed) known as the Non-Fossil Fuel Obligation (NFFOSo far as hydro is concerned these programmes have resulted in the building of 33 small plants to date with a total installed capacity of 20.8 MW. There is no programme for subsidising hydro greater than 5 MW installed. All the renewable energy programmes are funded by electricity consumers through the fossil fuel levy.. The Non-Fossil Fuel Obligation has been introduced in a series of phases, the fourth of which is approaching completion. The UK Government places strong emphasis on competition and the requirement for the price of new and renewable forms of energy to converge with the price produced from existing fossil fuel generators. The process is designed to stimulate cost reductions in renewable forms of energy. This has proved difficult for hydropower because of limited opportunities-for savings in capital costs. The main advantage the system has is the prospect of a guaranteed fixed price contract for a period of 15 years, with a 5 year planning window. This policy has stimulated the development of small-scale hydro in the UK and provided a mechanism for developing other forms of renewable energy.

7.3 Government Policy and Licensing

The water resources in England and Wales are controlled by the Environment Agency (EA), a recently established quasi- governmental body which has subsumed the duties of the previous National Rivers Authority. The EA issues licences for all "abstractions" from water courses.

7.4 Present situation

Hydropower provides only about 2% of the total electricity consumption of the United Kingdom Total installed capacity (excluding pumped storage sites) remained approximately the same over the last ten years, at about 1,400 MW But because of the gradual increase in total generating capacity, currently about 60,000 MW, its share is gradually decreasing. There is however a definite increase in small hydro, due to the NFFO, but this is too small to compensate for the declining market share.

7.5 Conclusions

Most rivers and lakes in the UK are not completely unaccessible for eel and downstream migration in most rivers is in most rivers not affected by hydropower stations. In all, hydropower probably has little effect on the local eel stock. But numerous dams and weirs hamper upstream migration and affect in a negative way the utilization of potentially useful habitat.

8. NETHERLANDS

8.1 Eel habitat

Estuaries The rivers Rhine, Meuse and Schelde flow through the Netherlands into the North Sea, in an intertwining network of branchings and anastomosations. Therefore the estuarine area is hard to define. Brackish areas amount to 500 km2. The Waddensea is a lagoon-like area of 14,500 km2 . Because of the inflow of water from the river Rhine directly through lake IJsselmeer and indirectly via the coastal areas of the North Sea, it is brackish in nature. The brackish areas have been greatly reduced by land reclamation, often resulting in freshwater lakes, where the eel populations still strive.

Coastal lagoons No true lagoons exist.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 72

Coastal water Eel are taken as a by-catch in the fishery for brown shrimp. The shrimp fishery has been operated by numerous small boats operating within a few kilometres from the coast; surface area some 50,000 km2. The nature and extent of the habitat for eel in these areas is unknown; the eel caught might be migrating animals. There are no statistics on the eel catches, but the impression is of a steady decline in the past decade. Additionally, eel are caught as a by-catch in all fisheries on the North Sea. Traditionally, this profit is for the crew members, with the exclusion of the ship owner.

Rivers All surface waters in the Netherlands are connected to the rivers Rhine and Meuse. In the absence of major obstructions in the major waterways, virtually all may be considered to contain eel populations. Surface area of the main rivers is less than 1,000 km2.

Canals, polders and smaller lakes The accumulated length of ditches in all polders amounts to approximately 300,000 km, corresponding to about 3,000 km2. The surface area of all fresh water, excluding lake IJsselmeer, is approximately 15,800 km2. Eel populations exist in all, but downstream migration of eels from these polders is only possible via pumps.

Lake Ijsselmeer The surface area of lake IJsselmeer has declined from 34,000 km2 in 1932 when the lake was closed off from the Waddensea to its present area of 18,200 km2.

8.1.1 Areas of water inaccessible to eel due to chemical or physical obstruction

Lakes/canals There are no cases of lakes becoming completely inaccessible due to chemical or physical obstructions, but there is reasonable doubt with respect to the efficiency of recruits passing physical (water management) barriers.

Rivers No real obstructions.

Estuaries Many estuaries have been enclosed, but in most cases the immigration of elver has not been blocked completely. There is some evidence that the habitat in enclosed estuaries is more favourable to eel. This includes lake IJsselmeer.

8.2 Hydropower

A total of 0.1 TWh was produced by hydropower in 1998. Its is generated by 2 low-head hydropower stations in the river Meuse, 1 station in a river Rhine branch and 2 in tributaries of the Rhine and Meuse. There are plans for another 3 hydropower stations in the Meuse and for 1 small-hydro station in a tributary to the Meuse.

8.3 Dams

The Netherlands probably is the country with the most highly developed dike- and dam system in the world. The dikes have been build for flood control along the rivers and the North-Sea coast and for agricultural purposes in former wetlands and peatland (roughly 1/3 of the country is lowland area, often below sea level). Dams and weirs have been built for navigational purposes: in total 6 in the Meuse, in the Rhine-system and both rivers share 1 tidal dam. There is an unquantified but extremely numerous number of small dams, weirs and siphons in rural areas for agricultural purposes. And there are many small weirs in the small rivers. All the dams in the large rivers and many of them in the smaller rivers are provided with fishpasses or these have been planned. But large rural areas are hardly accessable for juvenile eels due to a number of obstructions and migration of silver eels from these areas is only possible via pumps. An inventory of these obstacles such has been made in Belgium is not available, but the situation in the Netherlands is comparable with Belgium or even worse.

8.4 Conclusions

The low amount of hydropower probably only affects a minor part of the local stock of the Netherlands. But specifically in the river Meuse and Rhine-branches it affects the local stocks from France, Belgium and Germany. The many dams and weirs affect the local stock severely in dispersion and development, except in cases where eels are restocked, and the migration of silver eels of the local stock is negatively affected by pumps.

73 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

9. BELGIUM

9.1 Migration obstructions

A very high number of migration barriers like sluices, weirs, dams, watermills, drainage pumps, siphons are located on Belgian water courses. Inventories are being made in most of the catchments in Flanders and Wallonia. In Flanders, at this moment 3 000 km on the 20 000 km waterways (15%) have been inventoriated. 1 050 potential migration barriers have been located and were included in a database related to an interactive website (http://vismigratie.instnat.be/). On the river Meuse Wallonia has 15 dams. At the moment 11 of them are equipped with fish ladders.

Presence of these obstructions are considered as one of the major causes of the fact that normally structured riverine eel populations in Flanders are restricted to areas nearby the sea.

The Benelux decree on the free migration of fish species in the hydrographic river basins of Benelux countries (1996) aimed to guarantee the free migration of fish species in all water courses by 2010 (Benelux 1996). In Flanders as well as in Wallonia high efforts are taken to build fish ladders or to resolve migration obstructions. At this moment in Flanders 24 (2.3%) barriers have been resolved recently, either by constructing fish passes or simply removing the barrier. The aim is to resolve all of them by 2010.

9.2 Yellow eel habitat

In addition to the low accessibility for a migrating species as the eel, yellow eel habitat has been heavily influenced by poor water quality and structural habitat degradation for many years. However considerable efforts have been made in water purification programs (especially in Flanders where a lot of rivers and brooks where for a long time not viable for eels due to poor water quality), resulting in a general increase in water quality. Projects aiming to ameliorate general habitat quality of water ecosystems (restoration of natural river banks, river basin management, restoring meanders) are believed to result in an increase of the eel habitat area.

9.3 Hydroelectric power stations

In contrast to the efforts taken for restoring habitats and enabling free migration of fish, some new initiatives are actually taken to install hydroelectric turbines on Belgian watercourses. This was triggered by the recent European legislation (European Directive 2001/77/EC) on the promotion of electricity produced from renewable sources in the internal electricity market, asking for more green power to be produced.

This is particularly the case in Flanders where, at the moment 7 hydropower turbines are active, but on 19 localities hydropower stations are in preparation. The possibilities to construct turbines on another 150 places is envisaged (Belpaire et al. 2002). These stations all have small capacity. Small capacity power stations ususally have no or minor energetic gain but may induce considerable ecological damage. Also in Wallonia hydropower stations have impacts on local eel populations, e.g. on the river Meuse 6 dams are equipped with turbines for producing hydroelectric energy. Only one of them is equipped with a fish deflector for downstream migrators.

Prognoses made for new plants in Flanders reveal an estimated immediate mortality of ±17% per site. The cumulative effect of a hypothetical situation with 5 consecutive hydroelectric power plants would result in a mortality rate of over 60% without taking into account the postponed mortality.

It is expected that these new initiatives will put new threats on silver eel escapement (Belpaire et al. 2002).

9.4 Conclusions

The hydropower plants in the river Meuse heavily reduce the downstream migrating silver eels from the French part of its drainage area. Except for the river Meuse system, the local eel stock in Belgium probably is not affected seriously by hydropower. But plans for building more hydropower stations are threatening the local eel stock. The many dams, weirs and siphons reduce the presence of eels in large parts of the country to practically nil except for places where eels are restocked. In the lowland areas the silver eels migrating downstream (to the sea) are negatively affected by pumps.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 74

10 FRANCE

10.1 Topography and Hydrology

France has a continental climate influenced by weather systems from the Atlantic Ocean and the Mediterranean Sea. Hence the precipitation is abundant but very unequally distributed over time and space. The mean annual volume of precipitation is 440 billion cubic metres of which 180 billion is usable. The variability is significant, and the minimum amount possible during a 30 year period is 60 billion cubic metres.

10.2 Eel habitat

10.2.1 Extent of water currently supporting eel stocks

Estuaries In spite of the lack of a current survey, the extent of French estuaries can be esitmated at about 1,000 km2. All these habitats are inhabited by eel at some stage. Nevertheless, there is no current monitoring or regular surveying at national scale, except for some estuaries where some scientific work is being carried out (Somme, Arguenon, Vilaine, Gironde, Adour).

Coastal lagoons Almost all coastal lagoons along the Mediterranean Sea support eel exploitation (350 km2). Most of these areas are accessible for eel, although the water and habitat quality has decreased in recent years because of human activities (sewage, refineries, water management, among others: Cavailles et Loste 1988).

Reclaimed marshes The reclaimed marshes on the western Atlantic coast are 2,800 km2 in extent. They contain about 240 km2 of water, and 40,000 km of ditches. All these areas are accessible to eel and provide for a wide range of habitats except when agricultural practices lead to chemical or physical obstruction.

Lakes The extent of natural lowland and mountain lakes can be estimated as 500 km2. Hydrodams and drinking water reservoirs cover about 96 km2. The 3,000 man-made ponds extend over 1,100 km2.

Rivers and canals.

Average width Total length Surface (m) (km) (km2) m km km2 35 11,800 413 10 25,530 255 1 88,000 88 0.5 150,000 75

Total 280,000 840

Most of these areas are accessible by eel, at least in the downstream reaches. However, major dams can obstruct the rivers more or less completely at various distances from the sea.

10.2.2 Areas inaccessible to eel due to chemical or physical obstruction.

There are no data about the accessibility of these hydrosystems, but most of the mountain lakes, among them the Léman Lake (200 km2 ), are inaccessible mainly because of dams and to a lesser degree because of natural obstructions.

Usually, all lowland lakes and ponds are colonised by eel. Some of them, like the Lake of Grand-Lieu (40–60 km2 according to the water levels) support intensive commercial fisheries.

No precise data are available on river systems. However, annual surveys conducted by Conseil Supérieur de la Pêche (CSP), the State Fisheries Agency, indicate a decreasing density of eel in rivers with increasing distance from the sea. This density decline can occur immediately within the estuaries (for example, the Vilaine) and to a greater degree

75 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

because of multipurpose dams along the system. A rough estimate shows that 10 to 20% of catchments are therefore inaccessible, especially in the Alps, the Pyrenees and the Massif Central.

There is no extensive geographical review on chemical barriers to eel in French freshwater waterbodies except in a few cases where eel may highlight biocontamination by xenobiotics.

10.2.3 History of Hydropower

Hydropower has been used for electricity production in France since the end of the 19th Century. The oldest installations have now been in existence for more than one hundred years. A hydropower construction programme had been planned just before the Second World War. The most active construction period was after the War, and it benefited from technological advances and from the nationalization of the electrical sector in 1946.

10.2.4 Capacity and Energy Production

The installed capacity increased from 5.0 GW in 1948, to 17.0 GW in 1975, to about 25.0 GW in 1995. Of the latter amount, Electricité de France (EDF) accounts for 23.0 GW produced by over 500 plants and 140 major dams. The other private or publicly owned producers (such as the SNCF, the French railways, and the Compagnie Nationale du Rhone (CNR), own approximately 1200 plants. The total energy production in 1995 was 69 TWh of which 68.3 TWh were produced by EDF

10.2.5 Hydropower Facilities

The various hydropower installations in France are characterized by great diversity. Run-of-the-river plants account for 52 %, plants associated with locks or weirs account for 21 %, and plants associated with lakes or reservoirs account for 27 %. There are several pumped storage plants including the Grandmaison site which has the same power as two nuclear power stations. There is one tidal power plant on the Rance estuary, it is presently the only one in the world, and its annual production is 0.54 TWh. The installed capacities range from a few hundred kW to almost 2000 MW.

10.2.6 Hydropower's Share of Overall Electricity Production

Hydropower produces 15 % of the total electricity production of EDF, and forms a natural complement to nuclear power stations.It is responsible for 50 % of the regulation required by load fluctuations in the network, and it allows EDF to match production to the amount required by consumers, by calling on the production facilities in increasing order of their marginal cost. The base load is provided by nuclear power stations and by run-of-the-river hydro plants.

10.2.7 Legal Structure

In France, a law dating from 1919 gives the Central Government the right to issue concessions for all hydroelectric production of more than 4.5 MVA of installed capacity. Installations of smaller capacity are authorized at the regional level,.

The nationalization of electrical energy which took place in 1946 gave EDF the monopoly for the production and transportation of electrical energy. Also, the electricity generated by independent producers (the private sector) must be bought by EDF at a rate close to its sales tariff, provided that it is produced in conformance with current laws and regulations.

The French Government makes no distinction between the various types of hydropower stations from the regulatory or fiscal incentive perspective. It does, however, facilitate access to European Union loans for small scale hydropower.

The law restricts the development of large scale hydro sites to EDF In fact, all such sites have now been developed.

More general laws related to water resources development and the environment were passed in 1993. They define the constraints imposed on different water users. The EDF installations must conform to this law, and when a concession is applied for or renewed this results in long and complex administrative procedures, which take from 4 to 6 years to complete. Environmental impact studies in greater or lesser depth have to be completed. In addition, the duration of the concession has been reduced from 75 to 30 years, which makes it more difficult to amortize the cost of the submission and of the studies. This problem reduces the competitiveness of hydropower in an increasingly capitalistic environment, and presents a barrier to potential investors.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 76

10.2.8 Present Situation

The number of favourable sites which have not yet been developed is very limited. This scarcity of economically feasible sites means that total hydropower installed capacity has changed little during the past decade. The last major project was the commissioning in 1986 of the pumped storage facility at Grandmaison of 1800 MW. The variations in installed capacity are due to some new equipment (Le Buech, Puylaurent) but above all to the oversizing or the improvements of performance. Consequently, the average energy production has remained stable (69 TWh). The coming years should see a marginal increase in capacity, due to the installation of some run-of-the-river plants by independent producers.

10.2.9 Future Prospects

Future hydropower investments will be multipurpose schemes. Project discussions will involve local representatives, different water users, and EDF A good example is the dam at Puylaurent, which was recently commissioned.

This project, which has an installed capacity of 3.2 MW and a storage capacity of 12 million cubic metres of water is remarkable for more than one reason:

• It has been planned since the very beginning for different water users: irrigation, tourism, hydropower, and maintenance of minimum flow.· It completes a system of hydropower installations consisting of 4 generating stations and 5 dams, and optimizes the overall production

• It used innovative construction techniques such as "hydroplus" fuse gates and use of fly ash in the concrete.

• It has an innovative financing package. The dam is the property of local governments, which have arranged for the financing. It is rented to EDF, which operates it. EDF makes payments which cover amortization of capital costs and interest. At the end of the contract, the dam will become the property of EDF In contrast, the hydropower generating station associated with the dam is the property of EN, which planned and built it.

Future hydropower projects will be similar to Puylaurent in that several stakeholders will be involved: communities, water users, and EDF The water resources will be equitably shared among the users and environmental protection imperatives will be adhered to because both the public and the authorities demand it.

10.3 Conclusions

The high number of hydropower stations probably has a serious effect on the local eel stock in France. Case studies have shown that there are many more dams and weirs than for hydroelectricity generation only. The dams and weirs reduce the colonization of favourable habitat by the eel and increase the effects of density dependent processes of the local stock. Diking of large coastal wetlands, mainly for agricultural purposes, has resulted in habitat loss for the local stock on a large scale.

11 PORTUGAL

11.1 Eel habitat

Estuaries Areas of estuaries were rarely available. The Tejo estuary, the largest in western Europe, measures 320 km2 and the Minho estuary is 5 km2. Potentially, all estuaries are inhabited by eel populations and their densities seem to depend on the level of pollution.

Coastal lagoons Various lagoons exist at a short distance from the coast, communicating with the sea permanently or, in certain seasons of the year, through natural or artificial entrances. Six lagoons of a total area of 225 ha are located within 6 km from the sea, on the coastline between Mira and Quiaios, in the central region. Another five coastal lagoons exist, including Lagoa de Óbidos, the largest in the country, with a length of 5.5 km and a maximum width of 1.5 km. Fishing activities take place in the larger lagoons where eel are caught by trap, long line and rod-and-line.

There are two important estuarine coastal lagoon systems in Portugal, the Ria Formosa with an area of 17 km2 at low spring tide and 84 km2 at high spring tide, and the Ria de Aveiro with an area of 43 km2, connected to the sea by a 470- m wide canal. Eel fisheries are well established but unquantified in Ria de Aveiro.

77 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

Lakes There are no substantial lakes in Portugal. Small mountain lakes in the Serra da Estrela (Seca, Redonda, Escura and Comprida, and the Manteigas lagoons) should be mentioned, but eel have never been recorded in them.

Coastal water Scuba divers report the presence of eel at breakwaters or jetties made of quarry rocks. There is no targeted fishery for eel, but they occur in by-catch.

Rivers/canals The 12 largest catchment areas total 79,742 km2, almost equal to the total surface area of Portugal. The total length of the 65 main rivers is about 6,000 km. The reservoir areas of the largest Portuguese dams cover 266 km2, but eel populations, if they occur at all, are very sparse.

Lakes/lagoons In general, coastal lagoons as typical habitats for eel populations, show signs of accelerated eutrophic processes due to the introduction of wastewater effluents.

Regular opening of the canals which communicate with the sea, dredging activities to avoid accumulation of sand, and the control of pollution, could lead to high densities of eel.

Rivers/canals Fishing resources and the distribution of eel have been affected by the construction of hydro dams, mainly since the 1950s. Besides a huge number of small dams, 100 large impoundments exist in Portugal. Some of them are equipped with fish ladders and other similar devices (Carrapatelo, Crestuma-Lever, Pocinho, Régua, and Valeira, located in the River Douro, Penide in the River Cávado and Touvedo in the River Lima). Their ineffectiveness, bad design and unsuitable or non-functioning fishpasses, are reasons why some of the migrating species have disappeared. In some rivers, on the downstream side of the impoundments and close to the sea, the eel is subject to intense fishing. Approximately 70% of total Portuguese catchment areas are inaccessible to glass eel.

11.2 Hydropower

Electricity generated by hydropower stations accounted for 33.4% of the total energy production in 1998 (www.iea.org).

11.3 Conclusions

Although large estuaries and coastal areas are unaffected by dams and hydropower stations, 70% of the total catchment areas of the Portuguese rivers are inaccessible for glass eels, mostly by dams and weirs. These affect the local stock seriously. The large rivers Duero an d Tajo originate in Spain and are void of eels there due to dams in Portugal.

12 SPAIN

12.1 Topography and Hydrology

Spain is a country with several major mountain ranges and is second highest in Europe after Switzerland in terms of the average height above sea level. It has a land area of more than 500,000 km2. Spain's relief is characterized by the Castilian Plateau, which is divided in two by the Central Mountain Range which separates the North or Douro Sub- plateau from the South Sub-plateau. The Plateau is bordered by the Galician Massif, the Mountains of Leon and the Cantabrian Range to the north, the Iberian Range to the east and Sierra Morena to the south.

External to the Plateau are the Pyrenees, the Coastal-Catalan Ranges and the Betic Ranges. Between the Plateau and the External Ranges are the depressions of the Ebro and Guadalquivir river basins.

The Hydrology of Spain is characterized by great differences in rainfall between northern Spain, with areas that receive more than 2000 mm/year and southern Spain with extremely dry zones in the south-east including areas with less than 200 mm/year.The average rainfall is 612 mm but with great variations from year to year and prolonged periods of drought which can last up to four consecutive years. The most important rivers are the Miflo, Douro, Tagus, Guadiana and Guadalquivir, which flow into the Atlantic, and the Ebro, Jucar and Segura which flow into the Mediterranean.

Spain has 53 km3 of storage behind large dams which regulates 40% of its river flow, varying from 71% in the Ebro river basin, to 11% in the basins on the Galicia coast (www.iea.org).

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 78

12.2 Eel habitat

Estuaries Dill (1993) lists 60 rivers of > 60 km in length. The estuaries of all 60 have eel populations.

Coastal lagoons The main coastal lagoons are on the Mediterranean coast: Albufera (Valencia) and Mar Menor (Murcia) and in the Balearic Islands: S’Albufera (Mallorca) and Albufera des Grau (Menorca). Eel stocks exist in all. Catch data are scarce and biased.

Eel and glass eel fishing take place in Albufera, having 20 km2 of lagoon plus 19 km2 of rice field. It is a very important ecological system. Mar Menor is a hypersaline lagoon (45–52%) with a surface of 132 km2. Eel fishing takes place from October to May, but there is no glass eel catch.

Lakes A large proportion of the lakes are in Asturias, in mountain places. In other regions, the lakes are not developed for fisheries and many of them are inaccessible to eel. In Astuarias, lakes are in mountain areas, over 1,000 m above sea level and most of them over 1,500 m. There are no references to fish populations in these lakes and data on extent are only available for the biggest (18, 7, 10, 12 and 8 ha, total 55 ha). There are no references to the presence of eel. In lakes communicating with the Narcea–Nalon basin it is unlikely that eel exist as their migration is blocked by hydrodams.

Coastal waters The Astuarias coast measures about 300 km and it is known by sport fishermen that the eel is a common species (but not appreciated) in several places, mainly inside ports and near river mouths. Eel is not exploited.

Rivers/canals Spain has seven main rivers with a total catchment area of 397,654 km2 . In addition, in the north of the country there are a great number of short rivers (mountains are very near the coast) very important in water discharge, and having a total catchment of 56,487 km2, of which 4,827 km2 are for the Narvea–Nalon basin. The Pyrenees rivers have a catchment of 16,600 km2. Dill (1993) gives the number of rivers and streams as 1,800 with a total length of permanent rivers of 72,000 km.

Estuaries There are no studies on eel populations in Spain in recent years, but it is clear that there are eel in almost all the estuaries. The exception could be some river mouths and estuaries with a high level of industrial contamination, such as several cases in the Basque country, or some rivers degraded by agricultural pollution or tourist use, as happens in some Mediterranean areas. In nearly all estuaries glass eel fishing takes place.

Coastal lagoons There are no obstructions to eel migration into coastal lagoons, though in all cases the surface area of the lagoons has been reduced by agricultural or tourist use. For this reason the water quality has deteriorated.

Lakes Access to the majority of lakes is prevented by their distribution at high levels in mountain ranges.

Rivers/canals (a) Northern rivers: There are several main rivers and a large number of small ones. The whole catchment is 40,252 km2 including the Narcea–Nalon basin, and the Mino (natural border between Spain and Portugal) has a catchment area of 16,235 km2 in Spain. In Asturias the Narcea–Nalon basin has a series of obstacles, mainly in the Nalon branch, that together with pollution from coal mines probably caused the disappearance of eel. In the Narcea branch eel is present in almost all zones of the river except in mountain areas.

(b) Mediterranean rivers: The main rivers which discharge to the Mediterranean Sea are:

River Length Catchment area including reservoirs km km2 Ebro 910 85,500 Jucar 498 42,989 Segura 325 18,870

There are also a number of small rivers for which no information is available.

79 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

In the Ebro, due to the presence of a number of important dams, the eel is present only in the delta and lower part. For Jucar and Segura, prolonged drought and important hydrological works have seriously degraded the rivers and it is possible that the eel has disappeared from most of the areas, being present only in estuaries or river mouths.

(c) Atlantic rivers: These are shared with Portugal in which the estuaries are situated. The rivers (other than the Mino) are:

River Length Catchment km in Spain km2 in Spain Duero 920 79,300 Tajo 1,007 55,769 Guadiana 78 48,556 Guadalquivir 657 54,970

The Duero and Tajo rise in Spain and meet the sea in Portugal. In both cases there are many big dams, the first within Portugal, and in Spanish territory the eel has disappeared. The Guadiana forms the southern border with Portugal. The estuary is shared by the two countries, a margin for each one. It is likely that dams have the same effect as in the Duero and Tajo. No information is available for the Guadalquivir.

Eel are present in all the northern rivers except in the biggest where it is possible that they have disappeared from the major parts in some cases due to dam constructions and in others due to pollution. In the Mediterranean rivers, eel have disappeared from the middle and higher parts due to dams and pollution (industrial, agricultural and tourist problems). In Atlantic rivers, the eel has disappeared from the Spanish zone mainly because of dam construction and eel populations are largely confined to the Portuguese zone. No information is available about the eel population in southern Spain, but they exist in the lower parts.

12.3 History of Hydropower

The first hydroelectric power plants in Spain were constructed at the end of the last century. In 1901 40% of the electric power plants existing in the country were of the hydroelectric type.

With the appearance of alternating current at the beginning of the 20th century it became possible to transport electricity over great distances and there was a large scale development of hydroelectric power plants. The construction of major hydroelectric works called for considerable economic resources and numerous private companies were created, several of which still exist..

During the twenties, Spanish hydro-policy was based on the objective of the integral exploitation of hydrographic basins. This approach led in the following decade to the beginning of the integral exploitation of the Douro basin, an operation which was concluded in the forties and served as a model for the development of the rest of the river basins in the peninsula.

The creation of a serie of public electricity companies at the end of the forties added to the efforts which had been made before that time by the private electricity companies and boosted hydroelectric development. Due to the great irregularity of precipitation patterns many dams were built for combined irrigation and hydroelectric use.

From the mid sixties, the growing number of electricity generating installations in Spain became increasingly based on fossil fuel thermal power plants and later on nuclear power plants, which meant a gradual decrease in the percentage of hydro power in the total installed. Nevertheless, major pumping stations were built and there has been a continuing refurbishment of existing hydroelectric power plants and construction of new plants.

12.4 Present situation (1996)

Total electricity production in Spain in 1996 was 175.6 TWh, 3.6 TWh were imported, mainly off-peak, and 2.5 TWh were exported, mainly in peak hours.

Net consumption was 153.7 TWh, .

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 80

There has been an important growth in hydroelectric production due to the high rainfall and a considerable increase in self-generators.

The total installed power was 48,7 MW (36.3% hydroelectric, 48.4% thermal and 15.3% nuclear). There are more than 900 hydropower plants of < 5 MW.

12.5 Hydroelectric Potential

The theoretical natural potential, also known as gross potential, is assessed for the peninsular Spain at 150 TWh, which water consumption uses would reduce to 138 TWh.

The technical potential has been assessed at approximately 64 TWh, of which close to 7 TWh corresponds to hydro plants of less than 5 MW. Of this total, some 30 TWh could be produced in installations in service, leaving an unharnessed potential of another 34 TWh, which more recent studies reduce to 24 TWh.

In 1992 UNESA prepared a Catalogue of Hydroelectric Uses, not including mini-hydro energy, for the preparation of the National Hydrological Plan, identifying a remaining technical potential of 17.3 TWh. From this catalogue it was concluded that there is an unharnessed 6 TWh which would be economically feasible, though in half of the cases there are considerable or serious difficulties due to conflicts with other water or land uses or because of environmental reasons. When the Irrigation Plan and the new National Hydrological Plan are available it will be possible to quantify the importance of these difficulties.

12.6 New Regulation of the Spanish Electricity Sector and Hydroelectricity

During 1996 the Ministry of Industry and Energy decided to modify the present legal framework of the electricity system in order to encourage greater liberalization and to assure competition between electricity companies, taking measures to guarantee a lower cost of electrical energy to the consumer. (Reduction of 3% in 1997 and further reductions in the forthcoming years).

Given the complexity of the electricity sector, before the Government presented the proposed legal modifications to Parliament, the Ministry of Industry and Energy wished to reach a basic agreement with the main parties in the Spanish electricity system. To this end, a Protocol was signed, defining the operational basis which will govern the functioning of the electricity system, the terms, measures and safeguards which must be put into practice during the transition period, and the basis for the payment of non-competitive activities.

12.7 Assessment of the Present and Future of Hydroelectric Energy

At present, the formation of the price of electrical energy is based on known standard costs. On the one hand, remuneration is given for investments (amortization and capital allowances) and on the other hand fixed and variable operating and maintenance costs are remunerated.

In the future, all income will be linked to production and to the marginal value of the energy in a competitive system, though income may also be obtained by the guaranteeing of power and for complementary services.

Costs in the period of transition towards a competitive environment have been allocated for existing installations. They will be remunerated during a certain period of time, and this also applies to existing hydroelectric installations.

12.8 Planning

It will be necessary to wait to see how the Protocol develops and how complementary services and the guaranteeing of power are to be remunerated, as well as analyzing new projects in the light of the new framework of payments in order to see which will be profitable and may be constructed and which will not. It will also be necessary to take into account the National Hydrological Plan, currently in preparation, along with the Irrigation Plan, and criteria for environmental aspects, priorities of usage, etc.

The previous forecast, according to the National Energy Plan PEN-91, was of 1000 Gwh of new hydro power in the period 1996-2000 and a further 580 Gwh in mini-hydro plants.

81 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

12.9 Mini-Hydro Power

At present there is a special production regime which covers cogeneration, waste incineration and the renewable energies, including mini-hydro plants of up to 10 MW. This regime is characterized by offering producers the possibility of selling all of their excess production to the national electricity system at a high price, currently for an indefinite time.

In the future this special regime will continue to exist. However, it is intended that the price of such excess energy will remain high only during the first years of service. When the installations have been amortized, or after a certain number of years, the price will be reduced to a value close to the marginal price plus a small bonus.

12.10 Conclusions

The local eel stock in Spain is heavily affected by dams and hydropower generation, partly in Portugal, but most in Spain.

13. ITALY

13.1 Eel habitat

Estuaries The estuaries of the main ( > 60 km total length) rivers number 29, to which about seven minor estuaries have to be added. Eel are probably present in most of them, but professional eel fisheries (yellow and silver) are confined to a small number. Glass eel ascent, on the contrary, has been ascertained by experimental fishing in many main and minor rivers, and professional glass eel fisheries take place in a higher number, and in many channel mouths as well, owing to the fact that this kind of fishery is often mobile, fry fishermen moving from site to site in different regions with equipped trucks to catch and transport glass eel, besides the juveniles of other eurhyhaline species.

Coastal lagoons Most of the yield of yellow and silver eel fisheries comes from extensive culture in the coastal lagoons which cover around 1,500 km2, of which approximately 610 km2 are exploited at present. Of the exploited area, about 300 km2 are located in the upper Adriatic and 120 in the Po delta, the rest being scattered in Apulia, Campania, Latium, Tuscany, Sicily and Sardinia (Ardizzone et al. 1988).

Lakes Large freshwater lakes, greater than 5 km2, in Italy number over 370 and have a total area of about 2,045 km2 (Dill 1990). The number of Italian lakes employed in fisheries is about 150, of which 107 have an area larger than 20 ha, providing fisheries for both cold- and warm-water species.

On the whole, the presence of eel is known, from literature or other sources, in about 1,200–1,300 km2, including all the large Alpine lakes of northern Italy and the volcanic lakes of central Italy. In some of them eel is the most important commercial species and yields are sustained by means of restocking. Furthermore, 300 km2 of hydro dams and 1.5 km2 of irrigation reservoirs are present.

Coastal waters It is believed that there is no significant population of marine-dwelling eel.

Rivers/canals The number of rivers in total is about 98, but main rivers, those greater than 60 km, number 45, 16 of which are tributaries to others. The total length is 7,782 km and water surface area 78 km2. The total catchment area amounts to 231,000 km2. To this total, about 30,000 km, about 30 km2, of artificial canals must be added. Some, such as those which link the Sesia, Ticino and Po rivers to supply irrigation water, aid in the distribution of fish, and certainly of eel.

Eel are present in most main and minor rivers, but production by riverine fisheries is probably negligible at present, as it is never mentioned in official statistics. Eel professional fisheries are practised in not more than four or five rivers, and concern only limited portions of rivers and some canals. Sport fishing could account for a certain amount of riverine catches.

Lakes Even if lakes exploited for fisheries are about 40% of the total, eel are probably present sporadically in most basins, owing to migrations through tributaries or to methodical or occasional restockings. An exception is perhaps the case of the small, deep, high altitude Alpine lakes.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 82

Rivers/canals As with the lakes, eel are probably present in all the Italian riverine system, but undoubtedly abundance progressively falls with increasing distance from the sea, owing to the presence of numerous dams, most of which are not supplied with fishways or have inadequate fishways, and are therefore impassable. It can be roughly estimated that 60% of the rivers are inaccessible to eel. To enhance stocks, rather than to sustain yields, restockings are carried out by Province or Regional Administrations. Eel are scarce or completely absent from the upper reaches of rivers and from an unspecified number of mountain torrents.

13.2 Hydropower

Hydroelectric production represents the most important source of renewable energy in Italy, corresponding to about a quarter of the total electricity produced. At the moment (1997), the production is about 52 TWh per year, of which 35 TWh per year is produced by ENEL (State owned company) and the rest by local utilities and others (private companies). In Italy, there are still non utilized resources accounting for a potential production of about 15 TWh per year.

13.2.1 The Regulated Electricity Market

The tariff policies of the electricity sector are regulated by a newly installed authority which reports to the Italian Ministry of Industry. It will establish electricity prices, according to criteria fixed by the Government, with the aim of:

• guaranteeing the promotion of competition and service efficiency • pursuing suitable quality of service standards while respecting economic and profitability considerations • defining a reliable and clear tariff system, which combines the economic and financial aims of users with aims of a social character

This new National Authority has a mandate for the control and regulation of the electricity sector, and will operate with complete independence of judgment. It will also carry out consultative and informative activities for the Government, and it even has the task of establishing, granting and putting into effect European Community norms.

13.2.2 Government planning and operation of hydropower resources

The Italian legislation in the field of electricity production has undergone major revisions and innovations during the eighties and nineties. It now offers private companies the possibility of building power plants. Previously, this was a prerogative reserved for ENEL and local utilities. Since 1982 there has been a progressive liberalization in the construction and operation of power plants. Capital funds and incentives are even offered to the producers.

The incentives for hydropower extend for a period of eight years and are based on the avoided cost of fuel and the kind of power plant which is proposed (reservoir, pondage, or run of the river).

13.2.3 Government Programmes for small and large hydropower

There is no distinction between low and high head installations in Government programmes except for administrative aspects: the low head installations are regulated by regional organizations and the high head ones by the Government.

13.2.4 Government Programmes which Impede or Discourage Hydropower Development

There are no Government dispositions which impede hydroelectric development, but the Government has recently produced authoritative environmental legislation, which is more restrictive than before. This indirectly produces negative effects on construction and re-powering programmes, except for the more profitable ones.

13.2.5 Government Adjudication Measures for Water Resources

The Public Administration grants funds for the study of the best way to exploit the available water resources.

The use of water resources for drinking and irrigation has priority over the production of hydroelectricity. However, different uses can coexist within a "multi-use" plan for the resources.

83 O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc

13.2.6 Hydroelectric production in Italy during the last 30 years

The first year of ENEL activity was in 1963, and at that time the company's electricity production was largely of the hydroelectric type. The maximum capacity was 8,985 MW and the energy production was 31.22 TWh. In 1995, the maximum capacity of hydroelectric power plants had reached 16,390 MW but yielded the same energy production as in 1963. These data reflect the difficulty of increasing hydroelectric energy production in Italy, which almost reached its saturation level in the early sixties. This explains the effort devoted to the development of large power plants, especially pumped storage facilities.

The maximum capacity of thermal power plants in 1963 was about 3,485 MW corresponding to 27% of the total installed capacity. They produced 11.967 TWh of electricity. In 1995 the maximum capacity of thermal power stations reached 37,118 MW corresponding to 70% of the total and a production 141.608 TWh which was 80% of the total.

13.2.7 Hydropower Projects in the Past Five Years

No hydroelectric project has recently had any particular effects on public opinion while possible negative reactions have only occurred in a local sphere.

13.3 Conclusions

There are extensive coastal lagoons in Italy in which probably most of the local eel stock is present. These are relatively unaffected by hydropower and by dams. But the dispersion of eels in the inland waters is seriously affected by many dams and weirs. Hydropower generation is well developed in Italy and affects the eels that do migrate upstream or have been restocked.

O:\ACFM\WGREPS\wgeel\Reports\2003\WGEEL02.doc 84