Dipturus batis, the largest of the European Rajiforms, was once common around the coasts of Britain (Brander, 1981). In recent years due to fishing pressure, its numbers have receded dramatically from the Irish Sea (Brander, 1981) and the North Sea (Walker and Hislop, 1998). Little is known about the life history characteristics of this species, therefore, to better understand its biological characteristics a study was done, mainly using the data collected by the Glasgow Museum Tagging study, but also with data collected by the author. The data was split into three groups, A, B and C after it became clear that bias due to fishing effort and local changes in bathymetry influenced the results as a whole to greatly to be analysed. Results showed a significant association between the weight and depth for female fish in all areas (A, B and C), with larger fish being caught in shallower depths, this correlation was strongest in area A (Spearmans Rank Correlation 2-tailed test, p=<0.01). The same correlation was found for all males in the study area as well as (p=<0.05) and there was a strong association in area A (p=<0.01). Using age at maturity as 11 years old (Brander, 1981, Ellis and Walker, 2005 and Little 1995), weight at maturity was estimated to be between 33-42kg for males and 69-96kg for females, these estimates are substantially higher than those quoted in the literature and suggest, especially for males, that either D. batis matures earlier, or is heavier than previously thought at maturation. It was found that the economic worth of D. batis in area C is higher than its commercial value and that charter boat trips annually bring £292,500 to the local economy.
Keywords: D. batis, Rajiformes, seasonal migration, maturation, depth, weight, economic worth.
My thanks go to: Dr Alastair Lyndon from Heriot Watt University who supervised this project, his continual enthusiasm, assistance and guidance throughout this study was invaluable and greatly appreciated. Richard Sutcliffe from Glasgow Museums for initially allowing me the use of the Glasgow Museum Tagging Study data, which he himself has collected since its initiation, but also for the continued help and assistance he gave during the project. SNH, SSMEI Sound of Mull Project and Argyll and Bute Council Local Biodiversity Action Plan for funding the project. Dr Jane Dodd for accepting me to undertake the research and for her continued enthusiasm, help and support throughout this study, it made the collection of data in Oban run much more smoothly. Thanks also to all staff at the Oban SNH office who’s patience and support was much appreciated. Dr Sarah Cunningham for all of her help and support from beginning to end, especially her guidance with GIS software Ronnie Campbell for all the boat time he provided for research trips. His continued help and support throughout the write up of this report was greatly appreciated. I would also like to thank the following people for their contribution to this report: Davy Holt, Professor Jon Side, Adrian Lauder, SSACN, all the anglers who have contributed to the Glasgow museum tagging project throughout its course
CONTENTS Table of Contents ABSTARCT ACKNOWLEDGEMENTS TABLE OF CONTENTS TABLE OF FIGURES AND TABLES ABREVIAITONS CHAPTER .1 INTRODUCTION
1.1 Background on D. batis 1.2 Distribution 1.3 Feeding 1.4 Migratory Behaviour 1.4.1 Seasonal depth migrations 1.4.2 Population movements and aggregations 1.5 Life History 1.6 Pressures 1.7 Potential for Re-populating
Page number i ii iii v vii 1
1 1 3 4 4 5 6 8 11
CHAPTER 2. MATERIALS AND METHODS
2.1 The Study Area 2.2 Data Collection 2.2.1 Collection of biological data 2.2.2 Extra Field Work undertaken towards this project 2.2.3 Measurements from landed fish 2.2.4 Collection of economic data 2.3 Data Analysis 2.3.1 Population segregation 2.3.2 Life history 2.3.3 Movement 2.3.4 Population estimates 2.3.5 Economic importance of D. batis 2.3.6 Value of skate compared to commercial market 2.3.7 Statistics
12 13 13 16 18 19 20 20 21 22 23 24 25 26
CHAPTER 3. RESULTS
3.1 Behavioural Ecology of D. batis iii
3.1.1General Observation of the population throughout the study area 3.1.2 Seasonal observations of the population throughout the study area 3.1.3 Areas A, B and C 3.1.4 Annual migration 3.1.5 Population estimates 3.1.6 Fishing mortality 3.2 Economic importance of D. batis
27 33 37 48 52 52 53
CHAPTER 4 DISCUSSION
4.1 Maturation 4.2.1 Population segregation 4.2.2 Areas A, B and C 4.3 Migratory Behaviour 4.4 Population Estimates 4.5 Economic Importance 4.6 Life History
58 61 63 64 66 68 69
CHAPTER 5. CONCLUSIONS
Conclusions from this study Further areas for Study
Appendix 1 Individual Growth rates Appendix 2 Growth rate Scatter plots Appendix 3 Spearmans Rank Results Appendix 4 Distance travelled by each fish Appendix 5 total number of fish caught each year Appendix 6 Tables used for population estimates Appendix 7 Logsheets used for data collection Appendix 8 Questionnaires Appendix 9 Weight Charts
78 79 81 82 83 84 88 89 91
Table of figures
Figure 1.1 Map of D. batis Distribution Figure 1.2 Relative numbers of eggs laid per species on an annual basis Figure 1.3 Increase of smaller species in commercial landings Figure 2.1 Map of the Study Area Figure 2.2 Photograph of the Jumbo Rototags Figure 2.3 Diagram of restrictive nature of Rototag Figure 2.4 Diagram of D. batis showing tagging site for Rototag Figure 2.5 Floy Dart Tag Figure 2.6 Needle Applicator for Floy Dart tag Figure 2.7 Diagram of D. batis showing tagging site for Dart Tag Figure 2.8 Fishing gear used for catching D. batis Figure 2.9 Diagram of D. batis showing Gaffing sites Figure 2.10 Diagram of D. batis showing measurements taken Figure 3.1 Map showing the distribution of both sexes throughout the study area Figure 3.2 Histogram showing the number of Males caught in each weight class Figure 3.3 Histogram showing the number of Females caught in each weight class Figure 3.4 Scatter Plot showing weight related depth distribution for males Figure 3.5 Scatter Plot showing weight related depth distribution for females Figure 3.6 Sex ratio over time Figure 3.7 Map showing the seasonal distribution of males Figure 3.8 Map showing the seasonal distribution of females Figure 3.9 Median depth of males and females Figure 3.10 Growth rate of male D. batis Figure 3.11 Growth rate of female D. batis Figure 3.12 Sex ratio in areas A, B and C Figure 3.13 Histogram showing numbers of males and females caught in area A Figure 3.14 Histogram showing numbers of males and females caught in area B Figure 3.15 Histogram showing numbers of males and females caught in area C Figures 3.16 Histograms showing numbers of males in each weight class for all areas Figures 3.16 Histograms showing numbers of females in each weight class for all areas Figure 3.18 Map showing depths fish were caught at Figure 3.19 Median seasonal depth values for males and females in Area A Figure 3.20 Median seasonal depth values for males and females in Area B Figure 3.21 Median seasonal depth values for males and females in Area C Figure 3.22 Seasonal weight related depth distribution for females in area A Figure 3.23 Seasonal weight related depth distribution for males in area A Figure 3.24 Seasonal weight related depth distribution for females in area B Figure 3.25 Seasonal weight related depth distribution for males in area B v 2 7 10 12 13 13 13 14 14 15 16 16 17 28 29 29 31 31 32 33 34 35 36 36 38 39 39 40 41 41 42 43 44 44 46 46 47 47
Figure 3.26 Map of longest migration routes Figure 3.27 Map showing localised travel of males Figure 3.28 Map showing localised travel of males Figure 3.29 Map showing localised travel of females Figure 3.30 Map showing localised travel of females Figure 3.31 Mean distance travelled by males and females Figure 3.32 Population estimates Figure 3.33 Fisheries mortality for D. batis Figure 3.34 Percentage of participants on D. batis angling trips compared to Highlands and Islands Figure 3.35 Percentage contribution of charter trips from Oban to Highlands and Islands watersports Industry Figure 3.36 Amount spent on the top four water sports in the Highlands and Islands Figure 3.37 Comparative value of fish to t he sport and commercial fishing industries Figure 4.1 Sigmoid growth curve Figure 4.2 Diagram of hypothetical movement Table 3.1 Estimated amount of money brought to the local economy by anglers Table 3.2 Estimated value of an individual fish Table 3.3 Estimated value of fish to area C
48 49 49 50 51 51 52 53 56 56 57 57 60 65 54 54 55
ABBREVIATIONS USED THROUGHOUT THIS REPORT:
CEFAS: Centre for Environment, Fisheries and Aquaculture Science FRS: Fisheries Research Service ICES: International Council for the Exploration of the Sea IUCN: International Union for the Conservation of Nature MarLIN: Marine Life Information Network: The website of the Marine Biological Association of the UK and Ireland SAH: Scottish Angling Homepage SNH: Scottish natural Heritage SSMEI: Scottish Sustainable Marine Environment Initiative SSACN: Scottish Sea Anglers Conservation Network
1.1 Background on D. batis The Common skate Dipturus batis (formerly Raja batis or Raia batis Linnaeus 1758) was once common around the coast of the British Isles (Brander, 1981 and Fowler & Cavanagh, 2005) but due to increased fishing efforts, both commercial and sport angling, their numbers have rapidly decreased (Brander, 1981). D. batis is a favoured sport fish and has been pursued by anglers for many years, this has led to its depletion around the coast of Orkney, where a productive sport fishery has reduced its numbers (Anon, 2000). The number of D. batis caught in commercial trawls began to decline in the 1920s and, after a brief recovery period during World War II, they had all but disappeared from North Sea trawls by the early 1980’s (Walker & Hislop, 1998). One of the main reasons that these industries have impacted D. batis so much is partly due to our lack of knowledge of the species. The species is currently on the IUCN red list as an endangered species throughout its range (IUCN, 2007) and the UK Biodiversity Action Plan (Anon, 1999) although neither of these offers it any protection in terms prevention of landing and killing. The aim of this project was to gain more insight into the life history characteristics and behaviour for the species as well as assessing the population around the Isle of Mull, this was accomplished by using the data set provided by Richard Sutcliffe at the Museum of Glasgow. This report also looks at the economic value of D. batis to the local economy to the Oban area.
1.2 Distribution D. batis is the largest of the European Rajiform (IUCN, 2007) with females reaching a length of 285cm (approx 120kg, weight estimated from weight chart in appendix 9) and males, 205cm (approx 75kg, weight estimated from weight chart in appendix 9) (Ellis and Walker, 2005) (length is measured from snout tip to tail tip and can be seen in Figure 2.10), although there is evidence from commercial catches suggesting that they may grow up to 180kg in weight. In common with many other species of elasmobranchs, D. batis is slow growing, late to mature and has relatively low fecundity (Brander, 1981). They can be found in the northeast region of the Atlantic, ranging from the Mediterranean near Madeira and northern Morocco up to the Barents Sea and across
the Atlantic to Iceland (MarLIN, 2008) (Figure 1.1). They are found at varying depths throughout their range but are most commonly found in water around 200m deep on the edge of continental shelves (IUCN, 2007), although in the Barents Sea in the north of their range, they have been found in water deeper than 300m, and were most prevalent at depths of 550-800m (Dolgov et al 2005b). A CEFAS trawl study around the British coast classed D. batis as a deep-water species, distributed along the outer continental shelf edge of the Celtic and northern North Sea (Ellis et al, 2005). However, this does not represent the distribution of the species fully as it is often taken by shore anglers, suggesting that the preferential depth for D. batis cannot be easily assigned, it has also been hypothesised that depth distribution may be linked to the life history of the species. They are tolerant to temperature variation as shown by Dolgov et al (2005a) in the Barents Sea, where it the species occupied a wide temperature range even within a small geographic region.
Figure 1.1: Map of the known distribution of D. batis throughout Europe. Range marked in orange fill. Adapted from ZeeInZicht.
1.3 Feeding It is recognised that D. batis has a varied predatory diet ranging from benthic feeding on crustacea and molluscs (Little 1995) to demersal species such as cod (Dolgov, 2005c). In a study on the stomach contents of different species of skate in the Barents Sea by Dolgov (2005c) it was shown that fish contributed approximately 70% of the diet for D. batis with young cod, haddock, halibut and long rough dab being prevalent. Fisheries waste was also a major constituent of their diet, contributing 25% by mass, Dolgov (2005c) noted that this could not be seen as an indicator of poor abundance of prey species, but was more indicative of the large amount of available fisheries waste in the area due to high fishing intensity. There are some noted morphological differences in tooth shape between the sexes, these have been noted and substantiated by numerous anglers. Males appear to have more cuspidate teeth associated more with pisciverous fish, while females have more molariform teeth, more in common with bottom feeding fish, these differences in tooth shape suggest different feeding strategies, which might in turn suggest different habitat selection due to different prey items. As mentioned, D. batis is known for having a varied diet, but the possibility that this might vary with sex has never been studied and the sex of the fish studied in the Barents Sea was either not looked at or not disclosed. If it is the case that there is a difference in the feeding regimes between the sexes, then we may expect to see a difference in behaviour, especially migratory. It is possible that males, relying on migratory fish species as prey, follow the migration routes of fish such as cod while females, having a more varied diet, are able to stay within a smaller hydrographical region, opportunistically feeding on the benthos and in the demersal water column. However, it is known that males do feed on the bottom as anglers use benthic targeting equipment, and dead bait for skate fishing and regularly catch males. Dolgov’s study looked at 39 specimens of D.batis over 6 years, not a large number, which is perhaps a reflection of decreasing population levels. There was no breakdown of results by size for D. batis in his study so no change in diet for size/age is noted; other species however which were caught in larger numbers did have results broken down into stomach contents by length. For the Thronback Ray (Raja Clavata) it can be seen that as the fish grows, its diet changes, smaller fish having a more benthic based diet, 70% consisting of largely polychaetes and benthic crustaceans these diet components not being present in larger fish, where the diet consists mainly of fish (Dolgov et al, 2005c). If we assume a similar behavioural change in D. batis we might expect a similar change in diet, although we may anticipate fish to 3
appear in stomach contents at an earlier age due to the larger size of D. batis allowing it to handle larger prey items at earlier. If this is true, then we would expect only larger fish to migrate to follow fish stocks, while smaller immature fish may remain in the same place, feeding on benthic invertebrates.
1.4 Migratory Behaviour 1.4.1 Seasonal depth migrations The annual movements of D. batis are unknown and previous tagging programmes have been aimed at trying to understand any patterns in the movement of adult fish. However, since it is mainly anglers undertaking the tagging, it tends to be the larger, adult fish that are targeted. The initial thinking was that, due to their shape, they remain in the same place throughout the year, although Stephen (1929) proposed the idea that skate of all sizes and sex undertake annual migration to deeper water throughout the winter, returning to the same shallower waters in the spring time, other species of Rajiform have been shown to make such annual migrations. One of the best examples of this is shown research carried out by Hunter et al (2005 a & b) focusing on the migratory behaviour of R. clavata. Electronic data tags measuring depth and temperature every 10 minutes were used, which revealed that this species made an annual depth migration (Hunter et al, 2005a). The animals moved from deeper water (20-35m) during the winter months (November –February) into shallower water (<20m) in spring and early summer months (March – June). This supported previous studies (Walker et al 1997) and the timing was in direct relationship to the R, clavata reproduction cycle (Holden, 1975). If we put forward the hypothesis that D. batis performs similar migrations to R. Clavata, then we would expect to see an annual movement of the species from deeper water (600-200m) during winter months to shallower water (200-50m) in summer months in direct relationship to its breeding cycle, which is thought to be between March and June (MarLIN, 2008). Hunter et al (2005b) also focused on the movements of a single mature female R. clavata. Using the information gathered by an electronic data tag and a tidal location method (Hunter et al, 2005b), a migration route for this individual could be plotted over 423 days. A clear pattern was found as the female moved to the shallower water within the Thames Estuary in spring through to late summer time, presumably to spawn, then to the deeper waters off the Dutch coast during autumn and winter months. It repeated this migration 1.5 times, being released in London and recaptured off the Dutch coast. Although this was only one animal and 4
therefore not representative of the whole species, it does highlight that annual migration can occur within the Rajiformes and that some individuals are likely to remain in a regional population. The migration work of Hunter et al, (2005a&b) on R. clavata could explain the noted recapture of D. batis specimens within the same area (Little, 1995), Although it is known that some of the population remain in shallower coastal waters during winter months as some specimens are caught (Little 1997&1998) and there is anecdotal evidence from anglers that they catch D. batis during the winter months, although few assumptions about the decline in caught and recaptured fish during the winter months can be made as this may be down to decreased fishing effort alone (Little, 1995).
1.4.2 Population movements and aggregations It is unknown if D. batis populations are closed or recruit from other regional populations; it could be that individuals of both sexes move between populations. If this inter-population recruitment does take place it could explain why some fish were caught further away from their original catch site, they may have been en-route to another breeding area; there have been no genetic studies to prove or disprove this though. Trawl surveys undertaken in the North Sea have shown that skates and rays can have reasonably discrete species distributions, with each species occupying a slightly different environment, often depth related (Walker and Hislop, 1998) which suggests that species congregate, either all year round situation or only during spawning. The Glasgow tagging study has previously shown some possible congregation areas for D. batis on the west of Mull (Anon, 2000) where animals seem to gather and then disperse in northerly and southerly movements. As well as species segregation between Rajiformes, it is also thought that there is a degree of segregation within the same species based on sex, size and maturity (Ebert et al, 2008). It is unclear as to what extent this segregation takes place and whether segregated groups all move within the same population area, or whether each group occupy different habitats, based mainly on feeding and predator avoidance. In support of different groups occupying different areas, anecdotal evidence from anglers report that in the Crinan basin, off the west coast of Scotland, there seems to be a population of larger juvenile males and few if any females are caught. However, whilst the anglers claim these male fish to be larger specimens, the angling gear used would very rarely catch smaller individuals so a full picture of the resident weight range is unclear. Some 5
trawl surveys have caught mostly immature fish (Walker and Hislop, 1998) suggesting possible congregation of this age group in this region, although the data are not clear enough to determine if this is a single sex congregation or not. It was also noted in a Barents Sea study that some long line catches could compromise 100% of D. batis and other times, no D. batis were caught (Dolgov, 2005b), further suggesting that the species moves in groups, there was no breakdown of these catches with regard to sex or size mentioned. D. batis has also been shown to travel individually; a trawl survey along the edge of the continental shelf off Norway caught all its specimens of D. batis singly (Dolgov et al, 2005a). All of these differing reports suggest that the distribution and segregation of D. batis is complicated and little is fully understood. However, it is likely that in shallower coastal waters, mature populations come together for spawning as suggested by Little (1995 and 1997). Using the data collected through the Glasgow tagging programme, his study produced an interesting pattern regarding the movements of males into and out of the area. The proportion of males caught increased in the spring, suggesting an increase in numbers and declined again in the early summer, then increased again in September/October (Little, 1997). There is evidence from other Rajiform species that as their population levels decrease, populations have aggregated in certain key areas, with fewer pressures allowing for increased survival, these are becoming known as essential fish habitats (Payne et al (2008). If D. batis follows this survival behaviour we would expect to see a population increase within an area, as well as an increase in the diversity of sex, size and maturity as groups come together. Currently there is not enough information to map out segregation within species accurately.
1.5 Life History Growth rates for D. batis have been shown to be 1-9kg per year for males and 4-13kg per year for females (Little, 1998) and they are generally thought to live to 50yrs of age (IUCN). Both sexes reach maturity at approximately 11 years old, with males being 125cm long (Fowler & Cavanagh, 2005) and females 180cm (Walker & Hislop, 1998). The number of eggs laid per female is dependent on her size but is estimated to be approximately 40 eggs (Brander 1981) which get deposited on the sea bed once every two years (Little 1995), which is low compared to other European species of Rajiform. In a study by Koop (2005) on five Raja species (Raja brachyura, R. clavata, 6
R. microocellata, R. montaggui and R. undulate) it was shown that the average fecundity of females was 117 on an annual basis.
Figure 1.2: Relative numbers of eggs laid per month of five species of rays in the ‘Ray Reef’ of Dolphinarium Harderwijk, 1993-2001. Raja brachyura, R. clavata, R. microocellata, R. montagui and R. undulate. Adapted from Koop (2005)
However, the skate studied were all smaller species than D. batis (R. brachyura is the largest of the group at 125cm long, less than half the size of D. batis) and lay smaller eggs. If the fecundity of D. batis is compared to a Rajiform of similar size such as Dipturus trachyderma (Roughskin Skate), from the southeastern Pacific Ocean, which grows up to 258cm long (D. batis 285cm) there is similarity in their fecundity with D. trachyderma laying between 28 – 68 eggs, depending on the size of the female. This is similar to the estimate per female for D. batis, supporting it as a reasonable estimate. The exact breeding time for D. batis is unknown and is hard to estimate because Rajiformes are able to mate many weeks in advance of the females laying their eggs. The females are able to store the sperm and then lay eggs throughout a specific season Figure 1.2 (Koop, 2005). In the case of D. batis is in the spring and summer months (Ellis and Walker, 2005 and MarLIN, 2008). If we assume that sex segregation occurs, then breeding can only occur when the sexes mingle, according to previous analysis of the Glasgow data, this would appear to be in spring time when the proportion of males increases, this assumes that the sexes do not come together in another area (Little, 1995). The eggs are 14.4-24.5cm long (Ellis and Walker, 2005) and the length of time before the eggs hatch varies from 2-5 months dependent on temperature (Little, 1995). 7
These life history characteristics suggest that instead of a ‘bloom’ of hatchling skate, there is a more constant recruitment throughout the spring and summer. A more in depth study of the spawning cycle of D. batis could confirm fecundity, mating, laying and hatching season as the timing of the reproduction cycle of different species of Rajiform can vary considerably. This is highlighted in a study by Koop (2005) in which the reproduction of different species of captive Rajiform where looked at. All species displayed a similar reproductive cycle, although the timing and duration of this cycle varied considerably with some species laying large numbers during April to June season while other species lay lower numbers but more consistently between November and July Figure 1.2. The young of D. batis hatch from the eggs at approximately 22cm long with a 13-15cm wingspan (Brander 1981). They have fully developed spinal columns and sturdy spines (Walker and Hisplop, 1998) the movement of these hatchling fish is unknown as they are too small to be targeted by sport anglers and capture by commercial fishing is unrecorded. If D. batis does segregate by sex and size, then we would expect to see a mass movement of young out of the area. It is thought that breeding, laying and hatching occurs within the Firth of Lorne and Sound of Mull due to the presence of mature males and females during the spring and summer months (Little 1998) and anecdotal evidence from commercial trawl skippers confirming D. batis egg cases are brought up in trawl nets.
1.6 Pressures Elasmobranchs are well known for their vulnerability to commercial fisheries, and of these, skates and rays are arguably the most sensitive (Dulvy et al, 2000). For the same level of mortality among Rajiform species, it is expected that the longer the maturation period (Jennings et al, 1998) and the larger the body size (Dulvy and Reynolds, 2001) the greater the population decline. This places D. batis as one of the most sensitive species of Rajiform as it is a strong K-strategist (Stevens et al, 2000), has the longest maturation period for any species of Rajifrom and is the largest species in Europe. Even the hatchlings of D. batis are vulnerable to commercial fishing due to their size and morphology (Brander 1981). It is also true that species in higher trophic levels are more susceptible to decline (Jennings et al, 1999) and the size of D. batis puts it at a higher level than other Rajifrom species due to the size of prey it can consume, including smaller species of Rajifrom (Dolgov et al, 2005c). Current estimates suggest that the 8
population of D. batis could be declining by as much as 35% a year (IUCN, 2007). This large depletion of the population probably leading to significant changes in distribution (Oddone and Vooren, 2005) as is with the case of R. clavata which appears to be aggregating in ‘Essential Fish Habitats’ (Payne et al 2008) these are areas were pressures are fewer and the population is more likely to survive. The history of D. batis rebuilding former populations is not promising; after the initial depletion of skate around Orkney the population has shown little sign of restabilising itself after 30 years (Anon, 2000). This may be down to populations of D. batis being isolated so there are very few if any resident mature fish to enable repopulation. Elasmobranch species with the most potential to rebuild populations are short lived, early maturing inshore coastal species (Stevens et al, 2000) such as smaller species of Rajifrom, which after exploitation could rapidly (compared to D. batis) rebuild population levels. One of the main problems is that the UK commercial fishing fleet do not log Raja catches as individual species, instead grouping all catches of Raja species under ‘Skate and Ray’ (Dulvy et al 2000). Not only does this make it impossible to set catch quotas for specific species (Agnew et al, 2000) but it also makes determining the impact a commercial fishery has on a particular species using landing data very inaccurate. D. batis, however, is vulnerable to commercial fishing (Stevens et al, 2000), and is often taken as by-catch from long line, benthic and demersal trawl fisheries (Agnew et al, 2000). All skates and rays have some commercial value and commercial vessels will retain them especially as the exclusion of one species from a fishery is not an easy task. There is currently only a Total Allowable Catch limit on skate and rays set in the North Sea but this has historically always been above the total landings for this group (Payne et al, 2008), rendering it a very ineffective management tool. It has been shown that in the Irish sea, while larger species of skate such as D. batis have reduced in number due to commercial pressure (Gallagher et al 2004), smaller species such as R. montagui (spotted ray) and L. naevus (cuckoo ray) have increased in number. This may be due to the reasons mentioned previously, with the smaller species being able to rebuild their population more rapidly than larger species. This change in community structure has also been noted in the North Sea (Gallagher et al 2004). The proportion of species being landed is constantly changing and whilst the landings of skate and rays as a whole may not be substantially declining, the proportion of D. batis may be declining, but this cannot be detected due to mixed species landing data (Figure 1.3). 9
Figure 1.3: Hypothetical situation: Increase in proportion of smaller species being landed masking the decline of larger species being landed
Another reason for the sudden increase in proportion of smaller skate could be due to a change in their life history. It has been shown that long term exploitation of elasmobranches can lead them to change their life history (Tanaka et al, 1990, Walker, 1999), most noticeably, their growth rates (Stevens et al, 2000) and age of maturity (Tanaka et al, 1990) which means that these individuals will have a greater chance of surviving to breeding age (Licandeo et al, 2007). As these are selective attributes which are partly inheritable, it is expected that the population will evolve over time (Stevens et al 2000). Walker (1999) showed that this was true of certain Raja species (although not as yet noted for D. batis), although even if larger species do change life history patterns, the evolutionary nature of these changes on the population will take longer to occur in more K-selective species such as D. batis putting them at a further competitive disadvantage. As mentioned previously, D. batis was once common around the coasts of the UK. However, even at the beginning of the 1900s it was clear that the population levels were rapidly declining. In a study by Rogers and Ellis (2000) survey trawl data from the Irish and North Seas (in which specimens were identified by species) from 1901-1907 and 1989-1997 was compared. Even in the 1901-1907 D. batis was absent from the North Sea and only had low numbers in the Irish Sea (maximum 1 per hour trawl time (Roger and Ellis, 2000)) showing the reaction of the population to early commercial 10
pressure. In the trawls conducted between 1989 and 1997, D. batis was absent from all trawls. This has led to D. batis being declared extinct from the Irish Sea (Brander, 1981) and numbers disappearing completely from North Sea trawls (Walker and Hislop, 1998 and Philippart, 1998). Trawl data for the Celtic Sea up until 2003 showed that D. batis was still prevalent in the area (Walker & Hislop, 1998). However, the trawl survey around the coast of Britain showed that D. batis was found in the northern North Sea and Celtic sea only, in water 84-271m deep (Ellis et al, 2005), a lot shallower then the fish in the Barents sea. It is important to note that during these trawls, there was very low occurrence of immature fish.
1.7 Potential for Re-populating For a species with such a late maturation age and a low fecundity such as D. batis, survival to maturation is the most important factor (Walker and Hislop, 1998). The response to exploitation showed by some species of Rajiformes of increased fecundity would do little to increase the population size of D. batis as it is so susceptible to immature animals being caught. Far more important for species survival is the number of individual females reaching maturity. However, in a species with such a slow maturation period the initial survival of adults is paramount for population growth. This was shown to be the case for D. laevis (Barndoor skate), another large Rajifrom populating the Western Atlantic (Frisk et al, 2002). It has been documented that the species is robust and can survive the trauma of commercial capture well (IUCN) so if commercial vessels can be encouraged to return D. batis, there is a good chance the fish will survive, dramatically reducing fisheries mortality for this species.
2. MATERIALS AND METHODS
2.1 The Study Area The data used in this thesis were collected around the Isle of Mull, which lies between 6º23’18” - 5º38’45” and 56º39’25” - 56º15’52” and is located km off the West coast of Scotland (Figure 2.1)
57 42E3 42E4
Passage of Tiree
Sound of Mull
Isle of Mull
Firth of Lorn
Figure 2.1: Study Area around the Isle of Mull, western Scotland, with ICES rectangle numbers marked for the three used for Fisheries mortality analysis. Also marked a the three Skate angling areas, A, B and C used to split the Glasgow data into geographic areas.
This area was chosen due to the availability of historical angling records dating back to 1975 collected by various anglers in the area and compiled by Richard Sutcliffe, (Sutcliffe, 1994) at Glasgow Museums who kindly made these data available through SNH and they comprise the majority of those used for further analysis presented here. There were two phases of skate tagging in the area (1974-1988 and 1988-present day), 12
originally initiated (in 1974) by Dr Deitrich Burkel as a result of a dramatic decline in population numbers around the Shetland and Orkney Isles (Little, unknown date).
2.2 Data Collection 2.2.1 Collection of biological data Project ID
Figure 2.2: Photograph of the Jumbo Rototags used in phase 1 of the Glasgow Museum Study. Picture courtesy of Richard Sutcliffe.
Figure 2.3: Diagram representing the Jumbo Rototags initially used during phase 1 of the tagging study. Source: http://www.ketchum.ca/imgs/livestock/p7/versa.gif. viewed on 12/08/2008.
Figure 2.4: Showing the tagging location on the skate for the Jumbo Rototags.
In phase 1 (1974 – 1988) Dalton Jumbo Rototags (Little, unknown) were used (Figure 2.2) (normally used for ear tagging livestock). However these proved to be ineffective because, firstly, the tags offered a large surface area which allowed fouling organisms to attach, rendering it unreadable, while secondly, the constricting nature of the tag restricted growth (Figure 2.3) in the area of the fish around the tag, causing sores. The build up of fouling matter on the tag, which included hard shelled organisms such as barnacles, along with the rotational movement of the tag also caused damages to the skin of the animals. (Little, 1995). The Jumbo Rototag was placed on the trailing edge of the wing of the animal on either side as shown in Figure 2.4.
Figure 2.5: Showing the design of the current Floy Dart Tag with the barbed anchor at the left end of the tag. Picture courtesy of Richard Sutcliffe
Figure 2.6: Showing the Needle Applicator and the insertion of the tag into the needle ready for application. Picture courtesy of Richard Sutcliffe
Figure 2.7: Showing the tagging areas on the fish for the new Flow Dart Tag. The body core surrounding the internal organs is outlined in Grey
Due to the unsuitably of the Rototags, they were replaced by Floy FT-1 Dart tags (Figure 2.5)(Little, unknown) in 1988. These tags have a single anchoring point, do not cause injury and offer very little surface area for fouling. A needle applicator (Figure 2.6) is used to insert the anchoring point beneath the skin but above the muscle. The changeover of tags also represented the end of phase 1 and the beginning of phase 2. At the time of writing, Floy dart tags are still in use, although they have undergone a design change by the manufacturer, shortening the tags and changing the colour from orange to yellow. Floy Dart Tags could not be secured in the area previously used, as the tissue here was thin and the area subject to a lot of movement during swimming, which had the potential to dislodge the tags. Consequently they were secured in the more muscular areas towards the centre of the body (Figure 2.7), but avoiding areas above the internal organs. These areas provide a more secure anchor point for the tags and also are subject to less movement, reducing the risk of the tags being lost. (pers. Comm. Davy Holt). When Fish were initially caught and if they were not already tagged a tag was anchored to the fish, if a Rototag was present, this was replaced with a dart tag. Log Sheets provided by Glasgow museum were filled in for each new tag used. The details recorded were: tag Number, date caught/tagged, time, length (inches), longitude and latitude, location in words, sex, wingspan (inches), weight (pounds), depth the fish was caught at (depth of location opposed to length of line out), tidal state (open to taggers interpretation), bait used, tagger/captor and comments (such as condition of fish, any unusual features or circumstances). If the fish was already tagged, then the same details were taken but instead of a new tag being fitted, the number form the existing tag was 15
recorded and the tag left in the fish. There are some circumstances where the tag may be replaced, if the existing tag is a Jumbo Rototag then this is replaced with a Floy Dart Tag or if the existing Dart tag is loose and likely to fall out then this should try to be reanchored or if the anchoring point is damaged then the tag is replaced.
2.2.2 Extra Field work undertaken towards this project
Swivel to main line
Line between split link and main line. 40cm
Size 10 hook
Line between split link and main line. 240cm
Swivel with split 15mm plastic pipe link approximately 30cm. Snap swivel tied on with cable tie for attachment of lead weight (1.5-2kg)
Figure 2.8: Fishing gear used to catch D. batis and prevent deep hooking and cause minimal damage in the situation where the main line snaps, the weight and majority of line being detached from the hook. Picture Courtesy of Ronnie Campbell.
Figure 2.9: Showing the gaffing sights on D. batis
Figure 2.10: Showing the length and Wingspan measurements on D. batis
For the Field work undertaken in this project the Floy Dart Tags (Figure 2.6), provided by Glasgow Museums, were used along with a needle application tool (Figure 2.7). Log sheets were created (see appendix 7) which recorded name of skipper, date, weather, start time of fishing, end time of fishing, time fish caught, longitude and latitude (in WGS1984) existing tag number, new tag number, sex, depth (of location), weight (pounds or kg), length (inches or cm), width (inches or cm), tidal state, bait and comments. Also included were two images representing the dorsal and ventral side of the fish to for the purpose of marking any areas of damage, this damage could be further described in the comments box. The addition of start and end time is to allow for future analysis of fishing time or effort as this is one thing that is currently missing from the Glasgow data set. These logsheets were given to all skippers in the area running D. batis targeted charters. Any completed log sheets were collected at the end of the sampling period. Sampling took place form the angling boat “Laura Dawn” skippered by Ronnie Campbell operating from the Oban Times Pier in Oban between the dates 06/06/2008 – 26/06/2008 (not fishing every day). Fishing trips started at 9am and lasted 8 hours until 5pm, the amount of time spent fishing for D. batis varied each day due to a number of external influences. Firstly, the weather was the deciding factor as to the location for fishing as some, although offering better fishing, were very exposed and were unfishable in rough weather so the travel distance between Oban times Pier and the fishing location varied considerably which affected time spent fishing. Also, the first part of the 17
trip was spent fishing for bait fish, usually mackerel. The length of time spent fishing for mackerel depended on the number of fish caught, needed a minimum of 20 (personal observation) to use as bait for D. batis and also if the anglers bought any bait on board with them such as frozen squid, salmon then not as many mackerel were needed. The fishing gear used can be seen in Figure 2.8, this was attached to a nylon main line (5060lb approx 500 metres long) run from good quality reel with harness links attached to a 50lb class rods, there were commonly 6 rods over the side of the boat, but this was condition dependent, in rougher weather were the boat was subject to greater movement, fewer rods were put out to prevent entanglement with each other and the anchor rope. Bait was routinely checked as other benthic and demersal organisms in the area removed the bait from the hook. The bait was changed upon every inspection to try and maximise the scent trail along the sea floor. When a Specimen of D. baits was caught, the fish was brought to the surface and landed on the boat using gaff hooks, one for smaller fish, two for larger specimens, through the muscular tissue on the wings of the fish (Figure 2.9) avoiding the main body cavity. This procedure has been shown to cause no lasting damage to the fish and is the currently accepted method of landing the fish (Sharktrust, 2004) although some boat anglers chose not to land the fish at all, preferring to take measurements over the side of the boat while the fish is still in the water.
2.2.3. Measurements from landed fish Once the fish was landed, the length of the fish from the snout tip to the tail tip and the wingspan, the longest possible width remaining perpendicular to the length axis (Figure 2.10) were measured. The weight was often estimated using weight tables (appendix 9) as the fish are often too big to be accurately weighted at sea and the vast majority of anglers do not wish to kill the fish to take it back to shore for an official weight to be taken. Weights, wingspans and lengths were collected in imperial units as the equipment the skippers use is in these units as in the angling community, fish are still described in pounds and the weight charts use length and wingspan units of inches to estimate weight (in pounds) As information on the relationship between length, wingspan and weight has gradually been compiled over the years weight tables have been created for each sex to allow anglers to accurately estimate the weight of their fish from the length and wingspan measurements, these tables can be viewed in appendix 9. Sex is determined by the presence of claspers in male fish (Hamlett, 1999) although this can be harder to 18
spot in smaller specimens where immature males have very underdeveloped claspers. The current practice for anglers is to try and limit the time the fish stays out the water to three minutes maximum (pers. comm. Ronnie Campbell, Davy Holt and Adrian Lauder) Readings of position and depth should be taken before the fish has been landed, during the time the angler is playing the fish on the line, this is often in excess of 20 minutes due to the length of line out, but these readings can be taken afterwards, before the boat moves and should ideally be taken directly after the fish has been released. During the course of the field work which compromised of 12 days at sea, a total of six fish were caught and tagged, three aboard the Laura Dawn.
2.2.4 Collection of Economic data As part of the project, it was important to find out how much the population of D. batis was worth to the local economy as not only do 5 charter boats from the area run regular trips to fish for the species, but the anglers paying the charter also contribute to the local economy. To try and find out the worth of the species locally a socio-economic survey was undertaken. Questionnaires were handed out to the charter boat skippers with questions aimed at finding out how much they relied on the population of D. batis for their income. The information they gave also gave insight into how many charter trips they ran per year and which months they ran fishing trips for D. batis. The skippers were also given a questionnaire for the anglers to fill in (examples of both questionnaires can be found in appendix 8). These questionnaires were aimed at trying to find out if anglers were staying in the area, and if so, which type of accommodation they used and the group size of the party (including non-anglers). It was also designed to establish the expenditure of the group in the area, in order to evaluate out how much each charter boat contributed to the local economy. Aboard the Laura Dawn, these questionnaires were often filled out by the author directly, while the skippers from other boats gave the questionnaire out at the beginning of the trip and collected them at the end. A modified version of the anglers questionnaire, aimed to target all anglers of D. batis including shore based anglers and anglers with private boats was given to the SSACN to pass on to their members and also to Davy Holt to distribute on the skate tagging forum, directed through the SAH. All filled in questionnaires were returned via email or fax to the author
2.3 Data analysis Before any analysis of the data took place, the Glasgow data fully updated and filtered for any errors, erroneous and missing data. They were then converted to metric units (metres, centimetres and kilograms) using the following conversion factors:
1lb = 0.453 592 37 kilogram and 1 inch = 2.54 cm
The majority of the data collected has been measured in imperial units, in all instances, the data has been converted to metric units (metres, centimetres and kilograms). All entries for longitude and latitude were converted to decimal degrees, the method for this varying depending upon the original form of entries. The majority of data were entered as degrees and decimal minutes DD° MM.MM” although some entries were in degrees, minutes and seconds DD°MM”SS’. To convert these into decimal degrees the following formulae were used: Decimal Degrees = Degrees + minutes/60 + seconds/3600 or Decimal Degrees = Degrees +MM.MM”/60 The calculations for these were carried out in Microsoft© Excel© 2003 spreadsheet.
2.3.1 Population Segregation In order to determine if there was any segregation within the population of D. batis in the Mull area, the data were split into two groups by sex and weight class which were then both compared against depth in Microsoft© Excel© 2003 and geographic location (determined by longitude and latitude) in ESRI ARCview. Initially, a Spearmans-rank correlation was used in the software SPSS to see if there was any realtionship between weight of fish and capture depth the fish was caught for each sex. Further splitting of the data into seasonal groups, winter, spring, summer and autumn, allowed for annual patterns and movements to be highlighted. Each season covered three months with winter grouping catches in November, December and January, Spring grouping February, March and April, Summer grouping May, June and July and Autumn grouping August, September and October. The months for each season were chosen selected based on three variables, namely months spent fishing by different anglers and the breeding cycle of other species of Rajifrom taken from the work carried out by Koop (2005) and are also linked to the breeding cycle of R. clavata as shown in the
studies by Hunter et al 2005 a&b (the data were analysed again in Microsft Excel and ARCview). The data collected by the Glasgow Museum project were mostly collected by anglers and as such cannot be considered random sampling, since each angler who contributed data probably have preferred fishing areas decided by their own personal fishing history in the area and limitations on where they moore and operate their boat from. Also, different anglers fish during different times of the year with charter skippers mainly fishing during summer and Autumn, whereas the pleasure anglers fish all year round. In order to distinguish any movement patterns displayed by the skate from patterns created due to angling effort, the data were split into three geographic locations based on angling activity. These areas were the Sound of Mull (around the mouth of Loch Aline), referred to as Area A, the Passage of Tiree, referred to as Area B and the Firth of Lorne below the Junction of the Sound of Mull and Loch Linnhe, referred to as Area C. These areas are shown on Figure 2.1. Although these areas do not contain all fish caught, they do represent the majority of catches of D. batis in the study area and contain the majority of the records with geographic location given. The variables sex and weight class were then compared to depth within these geographic constraints.
2.3.2 Life History It has been shown that other Species of Rajifrom have changed their life history characteristics in response to mortality pressure (Tanaka et al, 1990 and Walker et al, 1999). We wanted to know if this might be the case for D. batis in the study area. As the Glasgow Museum data has been collected over the last thirty years it was possible to compare historical information to more recent data. The first thing looked at was the proportion of males and females and how this altered over time. As anglers target both sexes equally when they are fishing, it was assumed that fishing effort for both sexes was equal as no evidence on geographic preference by either sex could be found and the sexes showed no preference to bait, so proportions of each sex should be representative of population. Recapture data within the same year was discounted as this might count the same individual more than once and could lead to false values. Sex ratios for each year were looked at within each area, as fishing within each area covered different time periods. Another variable which was looked at was growth rate of each sex of the species. Individuals captured more than once with weight data for each catch incidence 21
were included in this analysis. An average value of kilograms change per day was calculated by dividing weight change by days between captures. This was compared to growth rates from a smoothed set of data (method outlined below) as most species of elasmobranchs have different growth rates throughout the year (Kusher et al, 1992). To smooth the data out and discount large changes caused by annual growth patterns, changes caused by feeding regimes and the carrying of gametes by females, the total weight change per individual between initial capture and final recapture was divided by total day’s freedom to gain an average growth rate in kg/day. In both cases, the figure for the growth rate was paired with the initial starting weight. The equation of this line was marked using Excel and used in calculations to create Figures for an estimated growth curve. Assuming a start weight of 0.5kg the line equation from the growth rate – weight scatter plot was used to work out the growth rate in kg/day for this weight of fish. 0.5 was used as an initial start weight by adapting the weight charts in appendix 9 as this figure was added onto the original starting weight (0.5) to give the weight after one day. This process was repeated for 5475 days (15 years, assuming 365 days per year) to create growth data for each sex over a period exceeding the cited maturation period of 11 years (Brander, 1981). This process was repeated four times, twice for each sex using the smoothed and unsmoothed results.
2.3.3 Movement The second objective of this project was to try and ascertain if individuals of either sex undertook any migrations as displayed by individual R. clavata in the southern North Sea (Hunter et al, 2005b). The recapture data were used to test this. Any fish that had been caught more than twice i.e. had two recaptures after its initial tagging catch were spatially analysed using ESRI© ARCview, individuals again being analysed by sex categories. The distance between each capture location for the same individual was marked in km and for each fish a total distance between capture sites was calculated, this was used to see if there were any annual migration patterns for either sex and to see if there was a difference between the sexes for total distance between capture points. The literature on maturity size for both sexes of D. batis is unclear and unconfirmed, so there is no definite weight split to distinguish mature from immature fish, to overcome this, the data was split into weight classes of 10kg intervals - 0-10kg, 11-20kg, 21-30kg, 31-40kg, 41-50kg, 51-60kg, 61-70kg, 71-80kg, 81-90kg, 91-100kg and 101-110kg and compared to each other. 22
2.3.4 Population Estimates As there is no unit of fishing time or effort the only model which could be used to estimate the population was the Schnabel model (McCallum , 2000). This is similar to the Peterson method but deals with ongoing sampling as opposed to just two discrete sampling episodes. The present data were split up into sampling periods so that recaptures and total catches could be compared over a given time. Three sampling periods were selected as the basis of population estimates. For all sampling periods, the same assumptions were applied, these being that the population of D. batis was closed, the mortality rate was zero, the tagged animals were not more vulnerable to recapture, death and had no competitive disadvantage compared to untagged fish, there was complete mixing between tagged and untagged individuals and finally, that none of the tags were lost. All of the above assumptions will lead to overestimates of the population but will provide a starting point for further analyses and evaluation of restricting parameters such as mortality. Daily sampling period The first population estimate used all the recapture data except for recaptures occurring on the same day, but does include fish caught in consecutive days. This is perhaps too short a time to allow mixing between the tagged and untagged individuals and is likely to provide the lowest population estimate. Annual sampling period For the second population estimate, each year was treated as a discrete sampling episode, discounting recaptures of fish within the same year, as this discounted many of the recaptures, as many fish were caught within the same year, this was expected to give the highest population estimate out of the three methods applied. 30 Days sampling period The third method is aimed at giving an intermediate estimate by including many of the recaptures but trying to leave enough time for mixing of the tagged fish with the rest of the population. As we know so little about the movements and behaviour of D. batis it is hard to give an exact time period to allow the fish to mix within the population, a time period of thirty days was used to allow for mixing of the population. This time length was chosen as most recaptures within a month occur at the same point or within the immediate vicinity of the original catch point, while recaptures after one month show greater dispersal which implies that they have mixed with the rest of the population.
Fishing Mortality In order to gain some level of understanding into the mortality from commercial fishing in the area data provided by FRS in Aberdeen were used as it shows the total skate and ray catch for the ICES squares 42E3, 42E4 and 41E4 which cover the study area (Figure 2.1). The original population estimates from the above sections are for the whole area so to work out fishing mortality for this population estimate, the ICES data was added together for each year to get a total skate and ray landings for the whole study area. Previous studies on Rajifrom species have run survey trawls in the Irish, Celtic, Barents and North Sea (Dolgov et al, 2005b and Ellis and Walker, 2005) and have noted skate and ray landings by species. These studies gave catch levels for D. batis in the form of either percentage of total catch or average number of individuals caught per hour fishing. With the information from these studies, a range of hypothetical catch levels as percentages could be applied to the FRS data to see what the landings of D. batis from commercial fishing maybe, in kg of gutted D. batis. To convert this number into individuals proved difficult as the commercial fishing is indiscriminate regarding sex or size as all life stages of D. batis are vulnerable to being caught (Brander, 1981). The Glasgow Museum Study was unsuitable to provide an average weight for D. batis as the data is collected by anglers who target larger specimens. An average weight for D. batis was gained by calculating the mean value of 50% maximum male weight and 50% maximum female of fish caught in the area, this gave a mean weight of 88.495kg. The kg catch was divided by this number to get a hypothetical value for individuals caught by commercial fleets. This figure was felt to be too vague to apply to the population estimates so has been left on its own as estimated mortality.
2.3.5 Economic Importance of D. batis Although the study area focuses on the Isle of Mull and surrounding area, the only number of viable returns for the economic survey all came from Oban, so the results are focusing on the economic importance of D. batis to Oban. With the data collected from the economic survey, the following results were worked out using Microsoft Excel formula with Minitab used for all statistical analysis. Firstly an average based on mean values, as the data was parametric as shown in Minitab using the Anderson Darling test (P value = <0.05), was gained for Money spent in area and an average based on the median value, as the data was shown to be non-parametric (P value = >0.05), for money spent on fishing tackle in the area were added together to get 24
a total average spend figure. Then an average, based on mean values (P value <0.05) for group size was calculated and the total spend figure was divided by average group size to get average spend per person excluding charter boat costs. The combined total for chartered days was estimated to be three hundred between the four charter boats operating from Oban, this was estimated from the returns of the skipper questionnaires (appendix 8). This multiplied by the average spend per person per day gave a total contribution of the customers partaking in fishing charters targeting D. batis to the local economy per year. This was then compared to an economic report on tourism in the Highlands and Islands (George Street), within this report was a section on watersports with total money spent in the Highlands and Islands. The figures gained from the study could then be used to assess how much money skate fishing contributes this as a whole. The figure for average spent per person on charter was compared to other watersports, such as fishing, scuba diving and sailing, to see how much money people spend on each activity. To assess the value of the Charter boat industry in Oban an average cost of charter hire per day was estimated from the returns of the skipper questionnaires and personal observation. This value was divided by the average group size to calculate the average spend on charter hire per person per day. This figure could then be used to calculate the total amount spent on chartered trips per year assuming 300 days at sea total.
2.3.6 Value of Skate compared to commercial market The SSMEI for Mull project was interested in the average value of individual D. batis in area. A population estimate had been worked out earlier in the thesis with varying levels of mortality. The Highest population estimate, i.e. without any mortality added, would give the lowest value per individual so this was used as a benchmark to estimate an individual worth for D. batis. The total value of the charter boat industry was divided by the population estimate to give a value per individual figure. In order to compare this to current commercial values (which are given in £ per kg), the average weights as used in the fishing mortality analysis were used to work out the average worth per kg of D. batis to the Charter boat industry in Oban. This value was then compared to values taken from the internet.
2.3.7 Statistics The Anderson-Darling test for normal distribution and Levenes test for equal variances were used in Minitab to determine whether further testing should be parametric or nonparametric. In all instances in this study, data proved to be non-parametric. Spearmans Rank tests in SPSS were used to compare the association between variables. Variables used were depth and weight of fish (appendix 3). As all the variables looked at were non-parametric, median values have been used for averages opposed to mean values as they more accurately represent the average values of the variables.
3.1 Behavioural Ecology of D. batis 3.1.1 General Observations of the Population throughout the Study Area The observed distribution of D. batis within the study area (Figure 3.1) shows that concentrations of specimens were concentrated in three main areas, the Firth of Passage of Tiree, the Sound of Mull and the northern end of the Passage of Tiree. These areas are ringed in black in Figure 3.1. The sexes were distributed equally throughout this range with neither showing preference for a particular area. Male specimens appeared to be more closely grouped in the Passage of Tiree and were absent from the west coast of the Isle of Mull while females were more widely distributed, including a scattering along the west coast of the island. There is a noticeable peak in the number of males being caught between 40.01 and 60kg with the largest number of male fish caught belonged to the 40.01-50kg weight class as shown in Figure 3.2. There were very few fish caught in the 60.01-70kg class. The numbers of females caught from each weight class shows a bimodal distribution (Figure 3.3) with numbers of individuals peaking in the 10.01 – 20 kg and 70.01 – 80 kg weight classes. Very few fish heavier than 90.01 kg were caught.
Area B: Based on fishing effort
Area A: Based on fishing effort
Area C: Based on fishing effort
Figure 3.1: Map showing the distribution of both sexes of the D. batis population around the Isle of Mull. The main map shows all captures of D. batis and the lower, smaller maps show captures of males and females.
Figure 3.2: Histogram showing the number of male specimens caught in the study area for each weight class between 1975 and 2008. Each weight class has a 10kg interval and the maximum weight for adult male specimens of D. batis in the area was 74kg.
Figure 3.3: Histogram showing the number of female specimens caught in the study area for each weight class between 1975-2008. Each weight class has a 10kg interval and the maximum weight for adult female specimens of D. batis in the area was 102.06kg
Figure 3.4 shows the weight related depth distribution of male fish. There was a significant association between the weight of the fish and the depth it was caught (result after performing Spearmans Rank correlation 2-tailed p = <0.001). There are two distinct groupings at different depths, group I, between 25 and 75m and group II, between 75 and 145m (Figure 3.4). The association between weight and depth can be seen when looking at the weights of fish caught in each group, group I only had fish greater than 10kg caught in it while group II had fish between 2-70kg, encompassing nearly the full range of weights showing that while large fish (greater than 10kg) are caught at in all depths of water, smaller animals (less than 10kg) are only caught the deeper water range encompassed by group II. In group I, the depth distribution of fish between 40 and 60kg is greater (between 25-75m) than fish outside this weight range, which are all caught at a similar depth (50-60m), there is no similar pattern in group II with all weights of fish being distributed throughout the depth range (75-145m). The weight related depth distribution for female fish (Figure 3.5) showed no association between these two variables (result after performing Spearmans Rank correlation 2-tailed p = >0.001). There are once again two distinct s depth groups, again labelled group I, 25-75m and group II, between75-145m. In group I only female fish weighing between 7kg – 83kg were caught, while in group B, fish between 1.36102.6kg were caught, showing larger weight range in group II. It can also be seen that in group I that there is an even distribution of weights, with no weight being substantially more represented than another, while in group II there are two groups based on weight, one between 1.36-30kg and the other between 60-90kg, these two groups are ringed in blue on Figure 3.5.
-250.00 0.00 20.00 40.00 Weight (kgs) 60.00 80.00
Figure 3.4: Scatter plot showing the depth distribution of male fish in relation to weight. Depth groups are circled in black
-250.00 0.00 20.00 40.00 60.00 Weight (kgs) 80.00 100.00 120.00
Figure 3.5: Scatter plot showing the depth distribution of female in relation to weight. Black circles are showing depth groups and blue circles are highlighting the bimodal distribution of weight in depth group II.
The proportion of males within the study area appears to have declined throughout the tagging programme as shown in Figure 3.6, with the large drop between 1988 and 1990 where the proportion fell from 70% down to just over 20%. Before this large decline, the sex ratio was even, with fluctuations, around 50:50, sometimes with a higher proportion of male fish, sometimes, more females. After 1990, the proportion of males did not rise above 45% and is currently at its lowest recorded value of below 20%.
Figure 3.6: Showing the change in sex ratio over time. Dashed line showing the trend for the proportion of males
3.1.2 Seasonal observations of the population throughout the study area When the geographic location of males was plotted with seasonal variation (Figure 3.7), a pattern could be seen. The numbers of males caught in summer and autumn were significantly higher than the number of males caught in the spring and winter. It can also be seen that there is an absence of males from the Passage of Tiree during the winter and spring seasons, while a large number of males were caught in this area during summer and autumn.
Figure 3.7: Map showing seasonal distribution of males throughout the study area with Seasons defined as winter: November-January, spring: February-April, summer: May-July and autumn August -October.
When the geographic location of females was plotted with seasonal variation (Figure 3.8), it could be seen that the numbers of females caught in summer and autumn was significantly higher than the number of females caught in the spring and winter. Another 33
pattern that could be seen, although not as distinctly as with males, is the dramatic drop in numbers of females caught in the Passage of Tiree during the winter and spring seasons while a large number of females were caught in this area during summer and autumn.
Figure 3.8: Map showing seasonal distribution of females throughout the study with Seasons defined as winter: November-January, spring: February-April, summer: May-July and autumn August -October.
Figure 3.9 shows that both sexes displayed a change in median depth throughout the seasons, with both sexes being caught at a similar, deeper median depth in the winter and spring and a shallower median depth during the summer. The median depth for males was 118.5m deep in the winter and spring and 92m deep in the summer. Females showed a similar pattern with average winter and spring depths being around 122m
during the winter and spring months and 102m in the summer months. During winter, spring and summer, the average depth for females was lower than for the males with the largest difference in the summer season. The median depths for both sexes were the same for autumn and lie approximately halfway between the median spring and median summer depths.
Season 0.00 Winter Spring Summer Autumn
-40.00 Depth (m)
Figure 3.9: Showing the median depth of male and female fish throughout the study.
Figure 3.10: Line graph showing the estimated Growth Curves for Males using Smoothed and Unsmoothed data for the first 15 years of the animal’s life.
Figure 3.11: Line graph showing the estimated Growth Curves for Females using Smoothed and Unsmoothed data for the first 15 years of the animal’s life
Comparing the two estimated growth curves for males (Figure 3.10), the unsmoothed data curve shows a faster initial growth rate, which gives weight differences of up to 12.5kg between the ages of 4 and 9 between the two curves (smoothed and 36
unsmoothed). This shows signs of levelling after 40kg to give a difference between the two curves of less than 5kgs. The lessening of the difference between the curves is caused as the smoothed data curve continues to rise relatively steadily, displaying more linear progression, showing that after the age of six, males experience a higher growth rate. The smoothed curve shows only a very small decline in growth rate between the ages of 7 and 15 years. Looking at the two estimated growth curves for the females, smoothed and unsmoothed, there is a large difference between them which continues to get larger as the fish get older instead of converging like the male growth curves in Figure 3.10. The initial growth rates for females are very rapid, although this is not clear in Figure 3.11 due to the scale of the axis, the smoothed curve, showing the upper estimate, shows females reaching a maximum of 30kgs towards the end of their first year, while the estimated weight predicted by the lower, smoothed curve shows female weight to be approximately 18kgs after the end of 2 years. Assuming a maturation age of eleven years old (Brander, 1981) the maturity weight for males is 33-42kg and for females, maturity weigh would be 69-96kg.
3.1.3 Areas A, B and C As Figure 3.1 shows, there are three distinct clusters of records; this next section looks at these areas individually in an attempt to remove local variability from the analysis. Figure 2.1 in materials and methods shows the geographic split based on angling activity which are similar to the three groups distinguished in Figure 3.1, it is these three areas, A, B and C that are used in the following figures. When the sex ratio in each geographic area was compared, it was found that there was a higher proportion of females in Area A, 70% than in Areas B or C which both had similar proportions 55 and 52% respectively, this is clearly shown in Figure 3.12. These proportions were taken from total numbers of males and females caught in each area as a cumulative total from all years.
Figure 3.12: Column Graph showing the sex ratio in Areas A, B and C
The distribution of males and females in each weight class varies throughout the three study areas. Figure 3.13 for area A shows a large increase in the number of females caught in the two weight classes between 60-80kgs, with up to 50 fish per weight class caught. The weight classes show little variation between the numbers caught, ranging between 20 (0-20kgs) and 12 (40.04-50kg). Area A had the largest female fish caught in the study area, weighing 102.6kg. The distribution of males in area A shows a steady increase in numbers of fish caught up to the 50.01 – 60kg weight class, with no substantial increase in numbers between each weight class. There were no larger males than 60kg caught in this area. In area B (Figure 3.14) there is no substantial peak of numbers of females in any weight class such as apparent in Area A, with the largest number of females caught being 40 in the 20.01-30kg class. Males, however, show a large increase in numbers, up to 70 fish, in the 40.01-50kg and 50.01-60kg weight classes, this a substantial increase of almost 50 fish from the weight class below (30.01-40kg). Numbers of males caught in other weight classes are relatively low outside the two peak weight classes, with numbers of fish caught not rising above 22. Area C (Figure 3.15) shows large peaks in the number of fish caught for both sexes. With females, the largest numbers of fish, are caught in the 10.01-20kg weight class (26), this declines in subsequent weight class to a low of 2 fish caught in the 38
40.01-50kg and 50.01-60kg weight classes. The number of fish then rise again to a high of 12 individuals caught in the 70.01-80kg class. Males have a substantial rise in numbers in the 40.01-50kg weight class with 29 individuals being caught. This quickly drops over the next two weight classes.
40 Male Female
0 0-10 10.01- 20.01- 30.01- 40.01- 50.01- 60.01- 70.01- 80.01- 90.01- 10.0120 30 40 50 60 70 80 90 100 110 Weight class (kgs)
Figure 3.13: Histogram showing numbers of males and females caught from each weight class in area A within 10 kg weight classes
80 70 60 Number Caught 50 40 30 20 10 0 0-10 10.01- 20.01- 30.01- 40.01- 50.01- 60.01- 70.01- 80.01- 90.01- 10.0120 30 40 50 60 70 80 90 100 110 Weight Class (kgs) Male Female
Figure 3.14: Histogram showing numbers of males and females caught from each weight class in area B within 10kg weight classes
0 0-10 10.0120 20.0130 30.0140 40.0150 50.0160 60.0170 70.0180 80.0190 90.01100 10.01110
Weight Class (kgs)
Figure 3.15: Histogram showing numbers of males and females caught from each weight class in Area C within 10kg weight classes
Figure 3.16 shows the number of males caught per weight class from each area compared against each other. It can be seen that most fish were caught in area B. In all three areas there is a similar pattern of gradually increasing numbers of male fish caught in each weight class until the 40.01-50kg class, which contains the highest number of fish caught for all three areas. This is most noticeable in area B where the numbers rapidly increase. The numbers of males in this weight class in areas A and C do rise, but not as sharply as area B. In all three areas, the peak of the numbers is within either the 40.01-50kg class (areas B and C) or the 50.01-60kg class (area A). There is no clear pattern in weight class proportion displayed by females (Figure 3.17), all three areas do have some level of bimodal distribution but the highest representation varies between weight classes. In area A, the largest numbers of females were caught in the 70.01-80kg weight class, in area B 20.01-30kg and in area C 10.0120.01kg.
80 70 60 Number Caught 50 Area A 40 30 20 10 0 0-10 10.01-20 20.01-30 30.01-40 40.01-50 50.01-60 60.01-70 Weight Class (kgs) Area B Area C
Figure 3.16: Histogram showing numbers of males per weight class in Areas A, B and C
40 Area A 30 Area B Area C 20
0 0-10 10.01- 20.01- 30.01- 40.01- 50.01- 60.01- 70.01- 80.01- 90.01- 10.0120 30 40 50 60 70 80 90 100 110 Weight Class (kgs)
Figure 3.17: Histogram showing numbers of females per weight class in Areas A, B and C
The map in Figure 3.18 shows that most fish caught at less than 75m deep were caught in area B near in the Passage of Tiree apart from one individual caught from the shore in area A. Fish caught at more than 145m deep were only caught in Area C in the Firth of Passage of Tiree. Fish caught between 75m-145m deep were caught in areas A and C.
Figure 3.18: Map showing distribution of depths fish were caught at throughout Areas A, B and C.
Figure 3.19 shows the median depth for fish in area A, which display similar seasonal depth distribution to the fish caught throughout the whole study area, with both males and females moving to shallower water in the summer. There is a difference of 11m between the summer median depth for males and the summer median depth for females in area A. The median depths for autumn are very similar for males and females, around 107m, being deeper than the median summer depth for males, approximately 114m, and shallower than the median summer depth for females, approximately 103m. Winter and spring depths are similar for each sex and only vary by a maximum of 2.5 metres, with females being caught at a median depth of 116.5m and males, 119m. Both sexes were caught at the deepest median depth in the spring.
In area B, shown in Figure 3.20, there were no fish caught in the winter and only 1 female caught in spring. The median depth value for the female caught in the spring is a lot shallower than the median depth for females in the summer. The median catch depth for both males and females is the same in the summer, approximately 56.5m deep and in both cases, this rises to a shallower median depth in the autumn, females being caught at a median depth of approximately 54.75m and males 53.5m. This is the opposite pattern that we see in area A. In area C, the sexes display very different seasonal variation in the depth they were caught in, Figure 3.21. In areas A and B, although the actual median value differed, the seasonal pattern of depth caught in was the same for each sex between the seasons. Females show progressively shallow median depths throughout the year, with their shallowest median depth (111.25m) in autumn and the deepest in the winter (115.8), while males were caught at their shallowest median depth in the spring (112.8m), and their deepest during the autumn (116.28m). This is the exact opposite of the seasonal depths in Figure 3.19.
-90.00 Winter Spring Summer Autumn
-105.00 Male Female -110.00
Figure 3.19: Column graph showing the median depth values for males and females in area A
-44.00 Winter Spring Summer Autumn
-50.00 Male Female -52.00
Figure 3.20: Column graph showing the median depth values for males and females in area B
-108.00 Winter -109.00 Spring Summer Autumn
-112.00 Male Female -113.00
Figure 3.21: Column graph showing the median depth values for males and females in area C
Figure 3.22 shows a scatter plot of depth against weight with seasonal data for females in area A, the association between weight and depth was shown to be significant using Spearmans Rank, 2-tail correlation test, p=<0.01. The general trend shows larger fish, 50kg+, being caught in shallower water. This is especially noticeable during the summer, where larger animals are found in shallower water, this can be seen in Figure 44
3.22 ringed in black. The fish caught in the shallowest water is 70kg in weight, heavier fish were found in deeper water again. In winter, there are three distinct groups, the first being 7 animals up to 25kg in weight caught in approximately 120m of water, the second group, two individuals between 20 and 40kg, being caught in 76m of water and the third group, 5 individuals, between being caught in 115m of water. These three groups are ringed in blue on Figure 3.22. Most females are caught at the same depth (between 130-140m deep) in the spring with a few outlying results, mainly between 5080kg, in shallower water. Most females caught in the autumn were above 60 kg in weight and were caught at shallower depths than most fish caught in the spring and summer time. There was one small specimen caught in the autumn under 10kg at a depth of 87m, shallower than any of the larger females fish. Larger males, (35kg+) in area A, in common with females, were found in shallower water up until 55kg, after this weight, they are caught in deeper water as within the circle region on Figure 3.23, once again, the association between weight of male fish and the depth was found to be significant (Spearmans Rank 2-tail correlation test, p=<0.01). This is most apparent in the summer, where fish between 35-60kg were caught at shallower depths, no fish lighter than 35kg were caught in less than 100m of water. Two groups are seen in the spring, with a fish between 5 and 45kg all being caught at a depth of 120m while larger fish 50-55kg are caught between 90-120m (ringed in blue on Figure 3.23). Figures 3.24 and 3.25 show the seasonal weight related depth distribution for males (Figure 3.24) and females (Figure 3.25) in area B. There were no winter results for either sex and only 1 spring result for females. Looking at the female data for summer and autumn, we can see that smaller individuals, less than 40kg, are caught between 50-60m, heavier females, 40kg+, have a greater depth range between 28-72m. In the summer, these larger animals, 40kg+, were more prevalent at shallower depths, 50-30m deep. Only one individual weighing less than 40kg was caught at a depth of less than 50m. The autumn female data appears to follow a linear progression (the ranges of this are marked with black lines on Figure 3.24) with larger fish being found at progressively shallower depths. The association between weight of female fish in area B and the depth they were caught at was found to be significant to a 0.05 level (Spearmans rank correlation test, p=<0.05). This is not true for males, although it appears that some larger fish, up to 50kg, are found at shallower depths, heavier fish are found at increasingly deeper depths down to 60m. The depth range for summer increases with weight, with males above 40kg being caught between 30 and 75m, below 40kg, the fish 45
are concentrated between 50-60m. There was a significant association between weight and depth for females in area C also (p=<0.05) but the scatter plot for this area did not show this and so scatter plots fort area C have not been included.
Figure 3.22: Scatter plot showing seasonal weight related depth distribution for females in area A
Figure 3.23: Scatter plot showing seasonal weight related depth distribution for males in area A
0.00 -10.00 -20.00 -30.00 Depth (m) -40.00 -50.00 -60.00 -70.00 -80.00 0.00
Figure 3.24: Scatter plot showing seasonal weight related depth distribution for females in area B
0.00 -10.00 -20.00 -30.00 Depth (m) -40.00 -50.00 -60.00 -70.00 -80.00 0.00
40.00 Weight (kg)
Figure 3.25: Scatter plot showing seasonal weight related depth distribution for males in area B
3.1.4 Annual Migration Figure 3.26 shows the longest confirmed distance travelled by D. batis between catch sites. There have been longer distances recorded, but these are by commercial boats and their accuracy is dubious (Little, 1995) and so are not been presented here. Fish 1 and 3 where originally tagged in the Sound of Mull and then travelled south to the Firth of Clyde. Fish 2 was originally tagged in Ballycastle Bay in Rathlin Sound then move north to the Firth of Passage of Tiree where it was subsequently captured the following year. These distances are all in excess of 200km. The median distance travelled by male fish was 6613m and 746m for females (taken from total distance travelled by all fish recaptured more than once). It can be seen from these median values that males, on average, travelled greater distances than females.
Figure 3.26: Maps showing the longest recorded ‘anger to angler’ distances travelled by male and female fish. Fish 1 and 3 were caught in the Sound of Mull and then the Firth of Clyde, fish 2 was caught in Ballycastle bay, Ireland, then in the Firth of Lorn.
The two movement patterns shown in Figure 3.27 show that the fish were being caught within the same area on a routine basis. The pattern between catch sites is an approximate figure of eight shape, with fish moving away from the initial catch site, and then moving back towards it after subsequent catches. The maximum distance between recapture sites was 5km
Figure 3.27: map showing catch sites of two male D. batis in the Sound of Mull and the Firth of Lorne. The lines and arrows shown on the map do not show the movement path of the fish, the fish is likely to have moved beyond the pathway of the line. The lines show direction of movement between the points and help to provide a time series between catch numbers of each fish.
Figure 3.28: Map showing the catch sites of two male D. batis in the Sound of Mull. One specimen is marked with blue circles and blue lines, the other with blue squares and black lines. Again, the lines are not representing the exact movement of the fish, but represent the direction travelled and time series of recaptures.
Female fish show a similar pattern as male fish with individuals seeming to move away from their initial capture site, then move back towards it after the first re-capture. The fish in Figure 3.29 was shown moving south in the Firth of Passage of Tiree from its original capture site, before moving north into the Sound of Mull, between the two geographic areas C and A. This was a longer distance (20km) between re-capture sites than displayed by males in Figures 3.27 and 3.28. The two females in Figure 3.30 show similar catch site locations to the male fish caught within the same geographic location. The fish move away from their initial site of tagging, only to move back towards it after the first recapture. The fish marked by red circles shows a repeated movement pattern along the Sound showing an almost circular movement. Using median averages, the distance between catch sites was greater for males than it was for females as shown Figure 3.30. On average, the distance between catch sites or males was 6613m, while for females, this was 736m, this is a difference of 5877m (5.877km).
Figure 3.29: Map showing the catch sites of a female D. batis Between the Passage of Tiree and the Sound of Mull. The lines are not representing the exact movement of the fish, but represent the direction travelled and help represent the time progression of recaptures.
Figure 3.30: Map showing the movements of two female D. batis within the Sound of Mull. One is marked by red triangles, the other red squares. The lines are not representing the exact movement of the fish, but represent the direction travelled and help represent the time progression of recaptures.
6000 Average distnce between catch sites (m)
Figure 3.31: Median distance travelled overall between catch sites for male and females
3.1.5 Population Estimates Figure 3.32 shows population estimates for the area using different assumed sampling periods. All the estimates show an increasing population with highest and lowest estimates for 2008 of 3318 and 1961 respectively; a difference of 1356. The middle population estimate shown is 2341. All lines show a rapid increase in the population after 1990 with the population continuing to rise until 2008. The annual sampling estimate shows an initial population decrease after 1979, which is not apparent in the other estimates. The annual sampling estimate also shows a slight decline in numbers in 2004, which is the same time period as a year of ‘no growth’ for the estimate gained from the 30 day sampling period. All estimates show a decrease in population in 1985 with numbers dropping by a maximum 200 individuals.
3500 3000 2500 Number of Fish 2000 1500 1000 500 0 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 All Recaptures 30 Day mixing Annual
Figure 3.32: Line graph showing population estimates for the whole study area. Each line shows a population estimate assuming different study periods. The red line is an estimate assuming an annual sampling period. The green line assumes a 30 day sampling period which allows for mixing. The blue line is an estimate assuming a 1 day sampling period, using all records except those caught on the same day. All estimates are with the assumption of no mortality.
3.1.6 Fishing Mortality The general trend is for the number of landed D. batis to decrease (Figure 3.33). This is true of all assumed proportions of total landings. The highest number of D. batis landed
was in 1984 when 198 specimens were landed. This dropped by 62 the following year to 136. After 1984 the levels of D. batis landed did not rise above 85. There was another dip (down to 49) in numbers being landed in 1994; this, however, recovered in 1996 to 73 fish landed. The lowest estimate shows 0.166 fish landed per year in 2007 while the upper estimate is 16.65. This is substantially different from the lower estimate of 1.56 and 156 in 1980, almost by a factor of 10.
FIsheries Mortality for D. batis with different % of Total Catch Scenarios
160 Number of Individual fish 0.10% 110 0.50% 1% 60 5% 10% 10
19 80 19 82 19 84 19 86 19 88 19 90 19 92 19 94 19 96 19 98 20 00 20 02 20 04 20 06
Figure 3.33: Line graph showing the estimated landings of D. batis by the commercial fishing industry in the ICES rectangles 42E3, 42E4 and 41E4 assuming different proportions of total landings. Assuming an average weight of 88.495kg for D. batis
3.2 Economic Importance of D. batis From the results of the economic questionnaires it can be seen that charter trips for D. batis contribute £292,500 to the local economy every year (based on results for 2008). Of this, £135,000 goes to the charter boat industry and £157,500 goes to the other businesses in Oban. Each member of an angling party spends £141.53 per day in Oban, £65.2 of this goes towards charter hire, the rest is spent within Oban town on food and drink, fuel, accommodation, other activities and goods.
Table 3.1: Showing the estimated amount of money that the D. batis angling fishery brought to Oban n 2008. The table shows the breakdown of this figure and how it was achieved by showing average spent per person, number of people a year the fishery brings to Oban and how much of this money goes to the local economy. Amount in £'s Total average money spent per trip per group excluding charter 1195.83 Total average spent per group per day excluding charter Average spent per person per day excluding charter Average spent per person on charter Total spent in area per person per day Estimated number of people per year Total spent per day fishing per charter boat Total spent annually in area (300 days charter fishing) Contribution to local economy Charter boat industry Gross income 525.00 76.21 65.32 141.53 2066 975.00 292500.00 157500.00 135000.00
Using the median number from the population estimates of 2342 from the entire study area (this number excludes mortality and therefore is an overestimate; any value placed on the worth of an individual fish is likely to be an underestimate) we can see that a single specimen is worth £124.89 in total with a breakdown of worth to local economy and the charter boat industry shown in Table 3.2 An estimate of the worth of fish in Area C is also useful, as this is the area most frequented by the charter boats based in Oban (personal correspondence, Ronnie Campbell) and so the population of D. batis in this area is the target of most charter boat anglers. The population estimate for area C is 169 and, as the money spent in this area and on the charter boat industry is all from Oban, we can assume that at least 80% of this money is spent on targeting fish in Area C. The values for ‘total spent in the area’, ‘contribution to local economy’ and ‘charter boat income’ are used for the calculations of fish value in Area C, as seen in Table 3.3.
Table 3.2: Showing the value of each fish throughout the study area, broken down into value as a whole, value to local economy and value to charter boats. Value of individual fish £67.25 £57.64 £124.89
Value to local economy Value to Charter boat industry Total Value of fish
Table 3.3: Showing value for fish in area C, the area heavily targeted by Oban based charter boats assuming an estimated worth of 80% of total money spent. 0% of money spent in: Total Spend on Area C Contribution to local economy of Area C Contribution to charter boat of Area C Amount £'s 234000 126000 108000 Minimum Worth of individual fish £1384 £745 £639
When the number of people who come to Oban to fish for D. batis (2066) is compared to the number of people who participate in all water sports in the Highlands and Islands (165, 000( George Street Research)), it can be seen that 1.25% of these participate in D. batis charter trips from Oban. Of the total revenue spent in the Highlands and Islands by water sports participants’ 1.56% of that is spent in Oban by charter boat customers. On average, charter boat customers spend £26.03 per person per day more than do other water sport participants. The most money is spent on water activities. The money spent on each activity is shown in Figure 3.36 where D. batis angling trips are the third most expensive activity after Scuba diving and canoeing/kayaking.
Oban angling trips for D. batis
Total number of people participating in water sports in the Highlands and Islands
Figure 3.34: Percentage of watersports participants who fish for D. batis from Oban. Actual figure is 1.25% of total watersprt participants from the highlands and Islands.
Figure 3.35: The percentage of total revenue to area brought by anglers fishing for D. batis. Actual figures are 1.53%
Figure 3.36. Amount spent on each water based activity. Adapted from George Street Research
If the value of D. batis from area C and the entire Isle of Mull population are compared, it can be seen that fish in area C are much more valuable to the local economy than fish in the Isle of Mull population (including area C) (Figure 3.37). If this is then compared to the current market value for skate (currently £2.40 per kg) which includes all species because skate is sold as skate and is not species specific, it can be seen that although the total value of the population is less per kg than the current market value, the value of fish in area C (£13.84 per kg) is comparably much larger than the market value.
16 14 12 10
Price per kg
8 6 4 2 0 Value of f ish in £ per kg
Isle of Mull Frith of Lorn
Figure 3.37: Comparative value of D. batis. Comparing current market price for skate (Grimsby Fish market, 2008) (mixed box of flesh, so all species included) price per kg of Isle of mull stocks, using lowest population estimate and price per kg for fish in Area C base on population estimate for area C and assumed 80% spend of total revenue in area C.
4.1 Maturation Broadly speaking, a population of any species can be broken down into four main groups, mature males, mature females, immature males and immature females, so it is useful to first get growth parameters for these groups for D. batis so the population can be studied with reference to them as we may reasonably expect D. batis to segregate its population based on sex, size and maturity as it has been shown that other Rajifrom species do so (Ebert et al, 2008). The current literature gives the age of maturation in D. batis as 11years (Brander, 1981) with males being 125cm long (Fowler and Cavanagh, 2005) and females between 140cm (Walker and Heesen, 1996) and 180cm long (Walker and Hislop, 1998). Using the weight charts in appendix 9, we can estimate the weight range that these lengths give. For males, the estimated weight would be 12.7-16.33kg and females, the lowest estimate would be between 17.23-21.77kg and the upper estimate between 41.28-51.71kg. Using the estimated growth curves in Figures 3.10 and 3.11 which were taken from recapture data for animals caught more than twice, it can be seen that the estimated weight range for 11year old fish in both sexes is substantially higher than those estimated from lengths quoted in the literature for D. batis. The growth curves given in Figures 3.10 and 3.11 are liable to be overestimates as they take into account weight change caused by food in the gut in both sexes and the spawning cycle in females. If fish are initially caught with an empty stomach and subsequently caught just after feeding, there will be a large weight gain recorded which will subsequently give a large weight gain from which the growth curves are estimated from. This effect will be compensated to some extent by fish that were initially caught directly after feeding with a full stomach then subsequently caught with an empty stomach, causing the growth rate to be low if not negative, A similar situation occurs with spawning females as female D. batis produce up to 40 eggs a year (Brander, 1981), measuring up to 24.5cm long, assuming an average brood size of 20 eggs, and a weight range of 10-20 g per egg, this could result in an apparent weight change (likely negative) of 200-400g per fish (likely negative). The other variable which could affect the estimated growth rate is the use of the weight charts to estimate weight. The common practice is for anglers is to use the weight charts (personal correspondence Davy Holt, Ronnie Campbell and Adrian Lauder) as opposed to weighing the fish with scales at sea (which is in any case difficult on a moving boat). Apart from the 5% error 58
that the weight charts give, there is also error produced by inaccurate measuring of the fish, as in rough sea conditions, or when measuring the fish in the water, the accuracy of the measurements may be lessened, even an error of 2.54cm (1 inch) either way can lead to a difference in weight estimate of up to 11kg (estimated from female weight charts, appendix 9). This error can also be introduced by intermediate measurements (i.e. 30.5 inches), the number will be rounded up to the nearest whole number (personal observation), giving overestimates in weight although the anglers try to keep the measurements as accurate as possible, allowance for errors should be taken into account. These errors will overestimate the weight of the fish, in turn overestimating growth rate. These combined factors lead to an overestimate of weight gain per day, the smoothed growth curves were produced to reduce this overestimate, they do however reduce the growth rate of smaller fish as the growth rate is taken as an average over time and any early periods of rapid growth are reduced by subsequent slower growth. The growth curves for males (Figure 3.10) show these differences very clearly, with faster, logarithmic growth being displayed by the unsmoothed data and lower, more linear progression displayed by the smoothed. For the females however there is no
logarithmic progression for either estimated growth curves and the smoothed data curve shows more rapid growth than the unsmoothed curve. The reasons that the opposite of what we would expect occurred could be due to annual growth patterns displayed by females, if we look at the growth lines for each fish shown in appendix 1, we can see that females experience no growth between November and February, these periods of no growth would show up on the unsmoothed data giving zero growth rate values which would lower the average value causing the unsmoothed growth curve to be lower than the smoothed growth curve as the smoothed growth rates would be averaged against less zero values as the weight change is from a longer period and less likely to be zero or negative. Neither of the growth curves show the traditional sigmoid shape (Figure 4.1) but this is to be expected as D. batis is strongly K selective (Stevens et al 2000) investing resources into producing well developed hatchlings.
Figure 4.1: Sigmoid Growth curve.
This will remove the lower section of a typical sigmoid growth curve (ringed in black in Figure 4.1) as this occurs within the egg case. This is shown to be the case with other species of Rajifrom such as D. trachyderma (Licandeo et al, 2006) and Bathyraja parmifera (Alaska skate) (Matta and Gunderson, 2007) and the curve that was produced for the growth of these two species is similar in shape to the growth curves in Figure 3.10 (smoothed) and 3.11 (both) for D. batis suggesting that they are reasonable estimates. As the estimated weight at maturity taken from the growth curves is between 33-42kg for males and 69-96kg for females we can assume, due to the overestimating nature of the growth curves, that the lower value is closer to the true weight. This is further supported by the scatter plots of weight against growth rate in appendix 2 where after these estimated weights of maturity, animals started experiencing weight loss, 33kg for males and 64 kg for females, this suggests onset of maturity, as more energy gets put into gamete production and weight loss is not uncommon in mature animals, especially during breeding seasons (Ref?). These estimates are significantly different to the current literature which, compared to this study, appear to be underestimates, especially in the case of males suggesting that maturation for D .batis occurs at an earlier age, or that D. batis are heavier upon maturity than previously thought.
4.2.1 Population Segregation The number of animals caught in each weight class throughout the whole area shows that there is a peak in the numbers of males (Figure 3.2) in the weight classes 40.01-50 and 50.01-60kg. When compared to the estimated weight at maturity it can be seen that these two weight classes encompass putative mature fish, so it appears that a large proportion of all males caught in the study area are probably mature. The numbers of females in each weight class (Figure 3.3) shows bimodal distribution with a large number of immature females being caught that were between 10.01-30kg and 60.01-90 kg. The largest number of females are in the 60.01-70 and the 70.01-80kg classes which are mature fish when using the growth curves and allowing for the over estimate. The high proportions of mature males and females in the region suggest behaviour in common with other Rajifrom species whereby mature individuals move to shallower coastal waters during the breeding season (Walker et al, 1997 and Hunter et al 2005b). Using the growth curves it is possible to estimate the age of the fish in the 10.01-20kg class to be between 6 months and 2 years old which suggest they are a generation below the mature adults. If D. batis does move to shallower water to breed and lay eggs, then the eggs will hatch in the shallower water and it is possible that they remain in this depth and use it as a nursery until they are strong enough to move into deeper, offshore waters to mature as we know that D. batis does occur in deep water up to 800m (Dolgov et al 2005b) but there was no classification of the weight of the animals found at this depth so it isn’t possible to estimate how old they are and this can’t be used to support an offshore migration of smaller animals. A possible reason that there is no bimodal distribution of males could be explained by 2 year old male fish being a maximum of 10kg (using the lower estimate from the growth curves, Figure 3.10) and are too small to be caught by anglers in the area who target larger fish. The change in diet shown by R. clavata shows smaller animals consuming largely benthic polychates and amphipos, this diet would allow smaller fish to retreat to deeper waters and continue feeding while also avoiding predation. As the fish grow and become less vulnerable to predation, they also change their diet to more active prey, such as migratory fish, and they will move into shallower water where these fish are present. (Dolgov, 2005c). It is known that D. batis preys on Spurdog (Squalus acanthias) in the area (pers. comm. with local fisherman), this is too large a prey item for smaller D. batis to take and the population 61
of S. acanthias and the Lesser Spotted Dogfish (Scyliorhinus canicula), of which there are high numbers in the area, would compete with smaller skate to which they have similar diets, both feeding opportunistically on macrobenthic fauna including hermit crabs, cockles and whelks (Dolgov, 2005 and Lyle, 1983). These feeding patterns support the theory that young D. batis move into deeper water to feed and avoid competition and predation, and then move into shallower water in accordance with a change of diet and fill a predatory niche with little competition from other species. This is only a hypothesis and the results from this project cannot support this fully, but it is an area that would benefit from further research with some targeting of fish smaller than 10kg in weight as well as sampling of D. batis in deeper, offshore water. An initial look at the geographic distribution of both sexes of D. batis throughout the study area shows three main areas of clustering at the Passage of Tiree, the mouth of Loch Aline in the Sound of Mull and the Firth of LornLorn (Figure 3.1) and it would be easy to assume, when linked to the large proportion of mature fish that these are possible breeding sites for D. batis. However, with more investigation into fishing effort it is revealed that the clustering pattern is purely down to fishing effort throughout the area, with anglers focusing their fishing in these three regions. What is clear from the geographic distribution throughout the study region is that both male and female fish are prevalent throughout the area. Fishing effort is also responsible for the grouping at depths, group I and II, seen in Figures 3.4 and 3.5. By looking at the depth fish were caught at in each area, A, B and C it can be seen that all catches in less than 75m were in area B, less than 145m in area C and equal distribution of fish being caught between 75-145m between areas A and C (Figure 3.18). Each angler tends to fish within the same area i.e. anglers fishing area A will rarely fish in area B, they also fish at different times of the year which appears to be the cause of the variance in seasonal geographic distribution. The main angler who fished area B only fished the months of March to October with peak fishing times around June/July (personal correspondence, Brian Swinbanks) while the main angler fishing area A is often out all year (personal correspondence, Davy Holt) and anglers fishing area C are generally out all year but with effort focused between March and October (personal correspondence Adrian Lauder and Ronnie Campbell). This explains the absence of fish in winter from area B and the large numbers in summer and autumn; this also offers an explanation for the observed shallower median depth in these seasons (Figure 3.9). 62
The obvious biasing of the data by the methods used to collect it makes it unsuitable to draw any conclusions on population segregation regarding depth and geographic distribution as a whole. However, if we remove the variable of local bathymetry and fishing effort by splitting the data into areas where the angling effort has been focused so all fish caught are subject to the same effort and the same localised bathymetry, where depth is more likely to be by choice rather than by location, then actual depth changes displayed by the fish can be looked at. This was done for the fish in nominal areas A, B and C shown in Figure 2.1 with each area being analysed separately.
4.2.2 Areas A, B and C The median depth for each sex between the areas show that both sexes in area A are in deeper water during the winter and spring (November – April) and in shallow water during the summer and autumn (June – October), in area B there were no records of depth for winter and only one record for spring making it unrepresentative, the fish were at their deepest during summer, although no depth data were available for winter and spring so no comparison against these seasons can be made. In area C males and females appear to follow different movements with females being at their deepest in the winter, progressing to their shallowest depth in the autumn, males however are found in deep water during the winter, their shallowest depth in the spring, and progressively deeper water over the summer reaching their deepest depth in the autumn. These patterns on their own might suggest that area A could be a breeding and/or hatchery area as the fish are moving to shallower water later in the year in accordance with their thought breeding season (Fowler and Cavanagh, 2005). Area C, with the rise of animals to shallower depth in the spring might also be a breeding area from which the males then descend to deeper water again, while the females remain to lay eggs. In an attempt to support these hypotheses, the depth distribution was plotted against weight with each season on the same area plot, the scatter plots for area C showed no patterns of depth related to weight although, in general, the depth of fish appeared to be slightly shallower in summer seasons, this was supported by a significant association between the weight of female fish and the depth they were caught in (Spearmans rank, p=<0.05), there was no association for the male fish in the area, more reslts from area C need ot be recorde before any firm conclusions can be drawn. In area A there was a distinct cluster of females larger than 60kg (borderline mature fish) found in shallower depths during the 63
summer, the records for autumn only included one fish smaller than 60 kg which happened to be caught in shallow water, assuming this result to be unrepresentative then a similar pattern during the autumn of larger fish at shallower depths can be seen. For males, this pattern is even clearer, with no fish less than thirty five kilograms being caught in less than 100m of water. During all four seasons, but most noticeably in spring and summer, the larger male fish (mature animals of 35kg and above) were found at shallower depths, this is similar behaviour displayed by other Rajifrom species (Walker et al, 1997, hunter et al 2005a) during breeding seasons, supporting the hypothesis that area A could be a potential breeding or hatchery ground with mainly the larger, mature fish being found in the shallower depth, to strengthen this hypothesis, there was a significant association between weight of fish (both male and female) and the depth they were caught in (p=<0.01). Area B showed similar patterns to area A for both sexes, although only females showed a significant association (p=<0.05), with mature fish rising to shallower water during the summer and autumn, this was again more distinct with males with no male fish less than 35kg being found in water shallower than 50m. Although area B is locally shallower than area A, it might be supposed that it is not the actual depth of water that is important, but the movement from deeper to shallower water that matters, in order for mature adults to congregate away from immature individuals or groups. This is assuming that both areas A and b are used for breeding purposes and cannot be assumed to be true. There were no records for winter and only one for spring so no comparisons against these seasons can be drawn These are hypothesis based on the known behaviour of other species of Rajifrom and although are only suggested by the data in this report are worth consideration and further research. There is not enough information to draw any conclusions for area C.
4.3 Migratory Behaviour Other species of Rajifrom do undertake annual migrations moving into offshore waters during winter and summer months and moving back into coastal waters during the summer months this movement appears to be linked to the breeding cycle, with the population moving to coastal waters during the breeding season (Walker et al 1997 and Hunter et al 2005b), it would be reasonable to assume from this that the fish breed and lay eggs in shallower, coastal waters (Hunter et al, 2005a). The records of migratory behaviour for D. batis are few, all using the Glasgow museum data, the conclusions drawn from these papers were that individual D. batis are associated with a regional 64
population with a few individuals performing long migrations (Sutcliffe, 1994 and Little, 1995). The results gained from this study generally support these hypotheses as there are several examples of fish being caught repeatedly within a small geographic area, displaying recapture patterns that suggest confinement to a small area,. Caution must be used; however, when looking at the recapture sites as, although they show were the fish has been caught and were the fish has moved to, they do not show movement in between the catch sites, we may only be looking at a small section of a larger pattern. Take the example in Figure 4.2, if the fish normally performs the migration shown by the solid black line but we only sample it within the boxed section, we might assume that the fish only moves within that area and the assumed route is shown by the dotted line. So although it cannot be said from the recapture data that the fish remains within one area throughout its life, it can be said that it returns to the same area on a regular basis.
Figure 4.2: Diagram to of hypothetical migration route with small sampling area in the centre of the route. Example of misleading recaptures.
The long migrations shown in Figure 3.26 prove that not all fish remain within a small geographic area. It cannot be assumed that the fish have moved from one population to another as we might only be sampling two points on a circular route, this is shown in Figure 4.2 where the two red circles represent sampling points, if we only catch a fish once at point X and once at point Z afterwards, it would be easy to assume that the fish had moved to point Z when in actual case the points are on a continual route. Of the fish that have been confirmed to make long journeys two, a male and a female, are recorded as having moved from the Isle of Mull to the Firth of Clyde in a southern movement, the other is a female moving from Ballycastle Bay (Ireland) to the Isle of Mull in a northerly movement. It is possible that these fish are performing similar routes between 65
the Isle of Mull and the Firth of Clyde but due the timing of catching, different conclusions for each fish could be drawn. The frequency and number of these migrations is unknown, partly due to the fact that fishing effort for D. batis is focused around Mull so the low record of long movements may be due to the lack of targeting outside of the study area rather than due to low numbers of fish moving. The other records of long migrations are from commercial vessels and there is a certain amount doubt about the location of the catch sites (Sutcliffe, 1994). A tagging programme in other areas, such as the Firth of Clyde where some of the Isle of Mull fish have moved to might increase our understanding of the movements of D. batis along the west coast of Scotland and would give insight into the regularity of these movements and whether or not they are one way migrations leading to recruitment into other populations or circular movements such as displayed by R. clavata in the Southern North Sea (Hunter et al 2005b).
4.4 Population Estimates Figure 3.32 suggest that the population of D. batis in the area is increasing. After talking to local fishermen, this does not seem unreasonable as some would agree that the population has been increasing over the last decade. Yet it would be unwise to assume from these figures that the population of D. batis in the area is increasing as there are other factors which can affect the population estimates, one of these is tagging effort. In the beginning of the project there were few anglers participating in the tagging programme, the common practice was to return the fish to sure for confirmation of weight landbased. This became less common and from the early 1990’s and more anglers began to participate in tagging in 1993, a concentrated tagging effort began in the Sound of Mull and in 1994, the Firth of LornLorn in; these years corresponds directly to the beginning of a rapid increase in population levels shown on Figure3.32, it can therefore be assumed that tagging effort is a major variable in the population estimates and it cannot be assumed that the increase in population size shown in Figure 3.32 is due solely to an increase in population. If the graph for population estimate shown in Figure 3.32 is compared to the graph in appendix 5 showing the number of fish caught each year discounting recaptures within the same year, it can be seen that a decrease in numbers caught directly relate to years with a decrease or zero increase in population, this further demonstrates that population estimates are heavily related to tagging effort. Little (1995) produce a population estimate of 60 -90 from the Glasgow 66
data, this was at the beginning of the increased tagging effort and so was likely to be an underestimate due to the lack of data being returned. We now know that the population is at least 110 fish as this was the maximum number of fish caught in one year and this would assume that 100% of the D. batis in the area had been landed. We can, however, use the population estimate in 2008 as a start point for future estimates and might prove a useful baseline to compare to any future research, but it must be emphasised that this is an overestimate and should not be used for any management plans, nor should it be assumed that there is a resilient population of D. batis in the area. If there is an increase in the population, this might not be solely attributable to recruitment through breeding, since there have been examples of Rajifroms congregating together in what is referred to as essential fish habitat (EFH) (Payne et al, 2005), it could be that the waters around the Isle of Mull are an EFH for D. batis. This would certainly explain why, in most other areas around the UK, D. batis is reported as being virtually extinct (Brander, 1981, Roger and Ellis, 2000). Although at initial glance this could partially explain the large increase in population, the bias put on the data by fishing effort does not permit this conclusion to be drawn directly, although it cannot be discounted and needs further research with data collection more suitable for population estimates. On a cautionary note, all estimates are overestimates, as the Schnabel model used does not add in any level of mortality. It also assumes that the population around the Isle of Mull is closed, we know that this is not the case as there are examples of individual animals both leaving and incoming to the area with up to 200km between catch sites (Figure 3.26). This shows that there is some degree of mixing outside the study area in this project although to an unknown degree. Another assumption for the model is that fish caught and tagged recover 100% and are at no competitive disadvantage to uncaught fish. This may not always be the case. One of the problems reported by some skate anglers is ‘deep hooking’ This is where the hook is taken down into the stomach and can either tear the inside of the stomach, resulting in death of the fish, or can remain impaled in the lining, causing feeding problems and weight loss. The main reason for deep hooking is the use of inappropriate fishing gear sometimes used by private boat angler or shore anglers, especially those not directly targeting D. batis. Deep hooking is sometimes a problem that cannot be avoided, due to the nature of skate landing on the bait and consuming it without performing a ‘run’ alerting the angler to ‘strike’ the fish, which would tend to set the hook in the mouth. Taking this into 67
account, it is fair to assume that some of the fish returned may be at a disadvantage to other fish and also may have a higher mortality rate(Committee on the Recreational Fisheries et al). There is a fishing code of conduct for D. batis which should be followed to minimise stress and damage to the animal, increasing its chances of survival after capture (available from Richard Sutcliffe at Glasgow Museum Resource Centre) The estimated fisheries mortalities all show the same pattern, a declining number of Rajifromes being landed. Due to the nature of Rajifrom landings all being grouped under ‘skates and rays’ the estimate for the proportion of this catch being D. batis works around assumed proportions of total catch and assumed average weight for D. batis which itself assumes that equal numbers from each weight class were caught. This makes the estimates liable to be inaccurate and not reliably representative of the actual D. batis take. It is likely that landings of D. batis have decreased, but to what extent is unknown without species specific landing data. Assuming a level of 10% of total catch, which is higher than any of the percentages in research trawls performed in other seas (Dolgov et a, 2005, Wlaker and Heesen, 1996 and Ellis and Walker, 2005), the pattern for most elasmobranch species in the southern North Sea is of declines in landings (Philippart, 1998), which is also true for the skate and ray landings from around Mull. Fishing effort in the area has not decreased substantially, and in fact the most days spent out of harbour were between 1994 and 1999. Reported landings of skate and rays do increase during this period, but not above levels pre-1986, suggesting that the numbers of skate and rays in the area are decreasing, most likely in response to fishing pressure, to which they are particularly vulnerable (Dulvy et al, 2000). Since D. batis is one of the most sensitive species to fishing pressure (Stevens et al, 2000) we can assume that there has been a proportionally greater drop in the numbers of D. batis being taken, compared to other Rajifroms.
4.5 Economic Importance The returns from the economic survey were not sufficient to conduct in-depth analysis into the economic value of D. batis to the area. Nevertheless, there were enough to make some general overviews as to how much money anglers brought to the area. For a relatively small industry, the charter trips bring a lot of people to the area, most of them travelling especially to the area for D. batis angling trips. Although the proportion of all water sport participants is only 1.25%, it must be remembered that skate angling is a specialist sport, with many anglers coming to Oban especially for it, staying for several 68
days and fishing for most of them. Proportionally, the amount of money spent by the anglers chartering boats is higher than the average spend by other water sport participants and contributes 1.53% of total income to the Highlands and Islands. It is hard to work out what percentage of money coming into Oban from water sports skate angling represents as there are no figures by town or even council area as to how much money is contributed by the local economy. The figures for contribution from skate anglers are estimates and would need more results to substantiate them; however, it is clear that the money brought into Oban by skate anglers is relatively large per capita. It appears from the results that fish in area C are far more valuable to the local economy than they are to the commercial industry, and it must also be remembered that skate caught by anglers and returned can be caught again, which doubles the value of that fish as two charter boat trips have enjoyed the benefit of catching that animal, which will in turn encourage anglers to return to the area to fish for D. batis again. Although the average estimated value of each fish throughout the whole study area is lower than the current commercial price per kg, it must be remembered that this value was obtained from a probable overestimated population level, which in turn, underestimates the value of the fish, so it is likely that the true value of D. batis is higher than that quoted in this here (£1.14 per kg)
4.6 Life History Although the Glasgow museum provides an excellent historical record of D. batis in the area, there were not enough records for every year with all the necessary data recorded to accurately assess any change in the life history of the population in response to pressure. Other species of elasmobranchs have been shown to do this (Tanaka et al, 1990), changing growth rates and age of maturity (Tanaka et al, 1990 and Stevens et al, 2000). Although it is possible that D. batis could change its age at maturity by increasing its growth rate and lessening its age of maturity, there is no evidence that this is the case. The results from this study strongly suggest that the age of maturity for D. batis around the Isle of Mull is 11 years old. This is the same age for maturity given 20 years ago (Brander, 1981) and cannot be seen to have altered. Due to the long life, late maturity and strong K selectivity of D. batis any changes to life history characteristics would be expected to occur over several generations so we would not expect any changes to happen quickly.
5.1 Conclusions from this study In conclusion, this investigation was able to show that immature and mature individuals of D. batis are present throughout the northern Passage of Tiree, the Sound of Mull and the Firth of Lorn. There was no observed geographic preference of either sex throughout the study area, although seasonal preference for one area was not attainable from the results due to the seasonal change in fishing effort in different areas.
There were clear trends showing annual movement of mature individuals of D. batis, of both sex, from deep water to shallow water in relation to the cited breeding cycle (Ellis and Walker, 2005) in certain areas which strongly suggested the preference of D. batis to breed and lay eggs in shallower water during the spring and summer in areas A and B. There was no preference to an actual depth displayed by either mature or immature individuals of either sex, this appeared to be related to local bathymetry. There is a strong suggestion that larger animals are found in shallower water, this association was shown to be significant for females in all areas, with the strongest association in area A (Spearmans Rank 2-tailed correlation test, p=<0.01) and males in area A and the study area as a whole (p=<0.01). Careful management of areas A and B might aid the rebuilding of the population around the Isle of Mull, this in turn, due to proven movements of D. batis, might help the population growth along the west coast of Scotland.
It is also clear that the Isle of Mull population is not closed, this can be seen by the long migrations of both sexes shown in figure 3.26. It is unknown to what extent these migrations occur and whether or not they are one way or cyclic.
This study also calls into question previous estimates for length and associated weight at maturity for D. batis. It is true that either D. batis matures at an earlier age than previously thought, or, that, the previous estimates for length at maturity from other papers are under estimates.
No clear conclusions can be drawn from the population estimates as the results are dependent onto many assumptions and are heavily influenced by the number of tag 70
returns. What is clear is that the tagging data needs more information, such as a record of time spent fishing for D. batis to enable a more in depth study into population levels is needed to determine the current status of the Isle of Mull population. As with population estimates, as we have no official record for the landings of D. batis it is unclear what impact the commercial industry has on the Isle of Mull population as this study assumes landing levels. Landing data recording individuals of D. batis is needed to look into the fisheries mortality.
Charter trips for D. batis bring a substantial amount of money to Oban, but it is unclear what proportion of total income this contributes to Oban itself, although to the Highlands and Islands as a whole, it is a relatively large amount for such a small industry. What is clear is that individual fish in the area are worth more to the angling industry than they are to the commercial market and, due to their ability to survive the trauma of netting, are worth returning to the sea upon capture.
5.2. Further Areas for Study The main conclusion drawn from this study is the need for further research into D. batis, not only in the waters around the Isle of Mull, but around the whole UK. If this species is to be preserved effectively, more knowledge on its behaviour is desperately needed to ensure its future, this study has opened the doors to many areas of potential research and I would suggest the following areas as targets for further study: • Genetic study. Looking into the mixing between regional populations of D. baits. This would indicate to what level populations mix and whether or not regional populations are closed or not. This is essential as closed populations are more vulnerable to fisheries mortality and are less likely to survive in exploited areas. • Study into the annual migration of both sexes of D. batis. Using electronic data tags or/and sonar tags to track the movement of individual skate more closely to see how common long migrations are and if they are cyclic by nature or one way. Also will show any annual migrations.
Study into the weight classes found in various areas of sea including offshore. This will further our understanding into the movements of young D. batis and will substantiate if they do indeed move into deeper water offshore to mature.
Sampling for D. batis throughout all waters around Mull. This will lead to better understanding of any geographic congregation areas for D. batis and will help in the allocation of breeding and hatchery areas for the species, allowing more effective management.
An in depth study into the breeding cycle of D. batis to confirm the months of breeding and egg laying for the species, again, to allow more effective management.
A more in depth study into the economic worth of D. batis to the Oban area, looking for contribution to local economy.
Collection of egg case data via trawl surveys and commercial trawlers. This will help us to understand the egg laying nature of D. batis and if females do congregate in a common hatchery, where these areas are.
Anonymous. (1999) Species Action Plan for Common skate (Raja batis) Originally published in: UK Biodiversity Group Tranche 2 Action Plans: Maritime species and habitats. Volume V. 47. Available from: http://www.ukbap.org.uk/UKPlans.aspx?ID=543#8. viewed on 16/05/2008 Anonymous (2000) Tagging of common skate (Dipturus batis) off the west coast of Scotland between 1974 and 2000. available from SAH: www.catchalot.co.uk/tagging/dec2000.htm viewed on 03/06/2008 Brander, K. (1981) Disappearance of common skate Raja batis from Irish Sea. Nature. 290. 48-49 Committee on the Review of Recreational Fisheries Survey Methods, National Research Council (U.S.). National Research Council (U.S.). Division on Earth and Life Studies (2006). Review of Recreational Fisheries Survey Methods. Published by National Academies Press, 2006 pp181. Dolgov, A.V., Drevetnyak, K.V. and Gusev, E.V (2005a) The Status of Skate Stocks in the Barents Sea. Journal of Northwest Atlantic Fisheries Science. 35. 249-260 Dolgov, A.V., Grekov, A.A., Shestopal, I.P and Sokolov, K.M. (2005b) By-catch of Skates in Trawl and Long-Line Fisheries in the Barents Sea. Journal of Northwest Atlantic Fisheries Science. 35. 357-366 Dolgov, A.V., (2005c) Feeding and Food Consumption by the Barents Sea Skates. Journal of Northwest Atlantic Fisheries Science. 35 495-503 Dulvy, N.K and Reynolds, J.D. (2002) Predicting Extinction Vulnerability in Skates. Conservation Biology. 16 no.2. 440-450 Ebert, D. A and Compagno, L.J.V. (2007) Biodiversity and systematics of skates (Chondrichtyes: Rajiformes: Rajoidei). Environmental Biology of Fishes. 80. 111124 Ebert, D.A and Sulikowski, J.A. (2007) Preface: Biology of skates. Environmental Biology of Fishes. 80. 107-110 Ebert, D.A, Compagno, L.J.V and Cowley, P.D. (2008) Aspects of the reproductive biology of skates (Chondrichtyes: Rajiforms: Rojoidei) from southern Africa. ICES journal of Marine Science. 65. 81-102
Ellis, J.R., Cruz-Martinez, A., Rackham, B.D., Rogers, S.I. (2004) The Distribution of Chondrichthyan Fishes around the British Isles and implications for Conservation. Journal of Northwest Atlantic Fisheries Science. 35. 195-213 Ellis, J.R., Dulvy, N.K., Jennings, S., Parker-Humphreys, M. and Rogers, S.I. (2005) Assessing the status of demersal elasmobranchs in UK waters: a review. Journal of the Marine Biological Association of the United Kingdom. 85. 1025-1047 Ellis, J. and Walker, P. Compiled by Fowler, S.L and Cavanagh, R.D. (2005) Grey, common or blue skate Diptururs batis (Linnaeus, 1758). Sharks, Rays and Chimeras. IUCN 333-335 Gallagher, M.J. Nolan, C.P and Jeal, F. (2004) Age, Growth and maturity of the Commercial Ray Species from the Irish Sea. Journal of Northwest Atlantic Fisheries Science. 35. 47-66 George Street Research and Jones Economics (2003/2004). Economic Impact and development opportunities for outdoor and Environment related recreation in the Highlands and Islands: Research Report: Watersports. Provided by Sona Magill at the Scottish Association of Marine Science. Grimsby Fish Market. Price List. Available from: http://www.grimsbyfishmarket.co.uk/frames/Prices.html. Viewed on 15/08/2008 Hamlett, W.C (1999) Sharks, Skates and Rays: The Biology of Elasmobranch Fishes. The Johns Hopkin University Press. Baltimore. 1 Hedges, Andrew. 2003 Convert Latitude/Longitude to Decimal. Direct link: http://andrew.hedges.name/experiments/convert_lat_long/ viewed on 26/06/2008 Holden, M.J. (1975) The Fecundity of Raja clavata in British waters. Journal de Conseil. 36. 110-118 Hunter, E., Buckley, A.A., Stewart, C. and Metcalfe, J.D. (2005a) Migratory behaviour of the thornback ray, Raja clavata, in the southern North Sea. Journal of the Marine Biological Association of the United Kingdom. 85. 1095-1105 Hunter, E., Buckley, A.A., Stewart, C. and Metcalfe, J.D. (2005b) Repeated seasonal migration by a thornback ray in the Southern North Sea. Journal of the Marine Biological Association of the United Kingdom. 85. 1199-1200 IUCN (2007) IUCN Red list of Threatened Species: Dipturus batis. [online] International Union for Conservation of Nature and Natural Resources. Available from: http://www.iucnredlist.org/search/details.php/39397/all viewed on 15/05/2008 74
Jennings, S., Reynolds, J.D and Mills, S.C. (1998) Life history correlates of ersponses to fisheries exploitation. Proc. Royal Society of London. 265. 333-339 Jennings, S., Greenstreet, S.P.R. and Reynolds, J.D. (1999) Structural change in an exploited fish community: a consequence of differential fishing effects on species with contrasting life histories. Journal of Animal Ecology. 68. 617-627 Koop, J.S., (2005) Reproduction of captive Raja spp. in the Dolfinarium Harderwijk. Journal of the Marine Biological Association of the United Kingdom. 85. 1201-1202 Kusher, D.I. Smith, S.E and Cailliet, G, M.(1992) Validation age and growth of the leopard shark, Triakis semifasciuta. Environmental Biology of Ifhses. 35. 187-203 Licandeo, R., Cerna, F. and Céspedes, R. (2007) Age, growth and reproduction of the roughskin skate, Dipturus trachyderma, from the southeastern Pacific. Ices Journal of Marine Science 64 141-148 Little, W. (1995) Common Skate and Tope: First Results of Glasgow Museum’s Tagging study. Glasgow Naturalist. 22, part 5.455-466 Little, W. (1997) Common Skate in the Sound of Mull. Glaucus. Vernal/Summer. 4243 Little, W. (1998) Tope and Skate Tagging off West Scotland: Part 2. Glaucus. Vernal/Summer. 36-38 Little, W. (date unknown) Tope & Common Skate tagging off the West of Scotland - an update. Available from SAH: http://www.catchalot.co.uk/tagging/tskate.htm viewed on 09/07/2008 Lyle, J. M. (1983) Food and feeding habits of the lesser spotted dogfish, Scyliorhinus canicula (L.), in Isle of Man waters. Journal of Fish Biology. 23, issue 6. 725-737 MarLIN, The Marine Biological Association. Direct link to Common skate (Dipturus batis) http://www.marlin.ac.uk/species/Dipturusbatis.htm. viewed on 10/05/2008 Matta, E.M and Gunderson, D.R. (2007) Age, growth, maturity and mortality of the Alaska skate, Bathyraja parmifera, ni the eastern Bering Sea. Environmental Biology of Fishes. 80. 309-323 McCallum, Hamish. (2000) Population Parameters: Estimation for Ecological Models. Blackwell Publishing. 87-95
Oddone, M.C. and Vooren, C.M. (2005) Reproductive behaviour of Atlantoraja cyclophora (Regan 1903)(Elasmobrnchii: Rajidae) off southern Brazil. ICES Journal of Marine Science. 62. 1095-1103 Payne, A., Cotter, J. and Potter, T. (2008) Management of elasmobranch fisheries in the North Atlantic. Advances in Fisheries Science. Blackwell Publishing, Oxford. Philippart, C. J. M. (1998) Long-term impact of bottom fisheries on several by-catch species of demersal fish and benthic invertebrates in the south-eastern North Sea. ICES Journal of Marine Science. 55. 342-352 Rogers, S.I and Ellis, J.R. (2000) Changes in the demersal fish assemblages of British coastal waters during the 20th century. ICES Journal of Marine Science. 57. 866-881 Sharktrust, (2004) Common Skate code of conduct. Available from Glasgow Museums. Stevens, J.D., Bonfil, R., Dulvy, N.K and Walker, P.A. (2000) The effects of fishing on sharks, rays and chimeras (chondricthyans), and the implications for marine ecosystems. ICES Journal of Marine Science. 57. 476-494 Stephen, A.C. (1929) Large skate from the Firth of Clyde. Scottish Naturalist. 175. 94 Sutcliffe, R. (1994) Twenty years of tagging common skate and tope off the west coast of Scotland. Shark, Skate and Ray Workshop. London. 15-16 February. 14-16 Tanaka, S., Cailliet, G.M. and Yudin, K.G. (1990) Difference in growth of the Blue Shark, Prionace Glauca: Technique or Population. Elasmobranchs as Living Resources: Advance in Biology, Ecology and Systematics and the Status of the Fisheries. U.S Department of Commerce, Washington. 90. 177-188 Walker, P.A. and Heessen, H.J.L. (1996) Long-term changes in ray populations in the North Sea. ICES Journal of Marine Science. 53. 1085-1093 Walker, P., Howlett, G. and Millner R. (1997) Distribution, movement and stock structure of three ray species in the North Sea and eastern English Channel. ICES Journal of Marine Science. 54. 797-808 Walker, P.A. and Hislop, J.R.G. (1998) Sensitive skates or resilient rays? Spatial and temporal shifts in ray species composition in the central and north-western North Sea between 1930 and the present day. ICES Journal of Marine Science. 55. 392-402
Walker, P.A. (1999) Fleeting Images Dynamics of North Sea Ray Populations. PhD Thesis. ZeeInZicht Digitale Encyclopedie. Direct Link: http://www.zeeinzicht.nl/vleet/index.php?item=blue+skate. Viewed on 15/06/2008
Appendix 1. Annual Growth patterns of individual fish
Figure A.1: Individual growth curves for females showing annual growth patterns.
Figure A.2: individual growth curves for males showing annual growth patterns
Appendix 2: Growth rate versus Weight Scatter Plots
0.15 y = -0.0688Ln(x) + 0.306 0.1
0 0 -0.05 10 20 30 40 50 60 70
Figure A.3: Scatter plot for weight against growth rate for males using smoothed data, showing logarithmic trend line and equation.
Growth rate (kg per day)
0.01 y = -0.0127Ln(x) + 0.0644 0 0 -0.01 10 20 30 40 50 60 70
-0.02 Weight (kg)
Figure A.4: Scatter plot for weight against growth rate for males using unsmoothed data, showing logarithmic trend line and equation.
0.12 0.1 0.08 Growth rate (kg per day) y = -0.0214Ln(x) + 0.1313 0.06 0.04 0.02 0 0 -0.02 -0.04 -0.06 Weight (kg) 20 40 60 80 100
Figure A.5: Scatter plot for weight against growth rate for females using smoothed data, showing logarithmic trend line and equation.
0.35 0.3 0.25 Grwoth rate (kg per day) 0.2 0.15 y = -0.0492Ln(x) + 0.2888 0.1 0.05 0 0 -0.05 -0.1 Weight(kg) 20 40 60 80 100 120
Figure A.6: Scatter plot for weight against growth rate for females using unsmoothed data, showing logarithmic trend line and equation.
Appendix 3: Spearmans rank results for Weight against Depth
Table A.1: Spearmans rank results for all males and females in study area. Significant association between weight and depth for males
Depth 1.000 . 829 -.065 .061 827 Depth 1.000 . 448 .179(**) .000 448 Weight -.065 .061 827 1.000 . 830 Weight .179(**) .000 448 1.000 . 468
Correlation Coefficient Sig. (2-tailed) N
Correlation Coefficient Sig. (2-tailed) N
Correlation Coefficient Sig. (2-tailed) N
Correlation Coefficient Sig. (2-tailed) N
Table A.2: Spearmans rank results for males and females within each are, A, B and C.
Appendix 4: Distance travelled by each fish, male and female
Table A.3: Total distance travelled by each fish re-caught more than 3 times. Median values shown at the bottom.
Sex Female 134093 0 0 5663 0 7859 100 150 0 0 0 0 16346 509 752 1904 250 2178 8757 11012 4178 4217 5744 5527 9947 317 8062 150 200 150 243330 100 100 250 4887 1415 740 100 32104 800 8501 100 253020 0 6613 Male 8530 5525 4458 0 9383 8543 0 8733 298 8561 7787 5768 224990 10744 1555 1295 895 10243 3079 11046 7458 4765
Distance travelled (metres)
Appendix 5: Total number of Fish caught each year within the study area.
19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03 20 05 20 07
Figure A.7: Number of fish caught discounting recaptures within the same year within the study area
Appendix 6: Tables used for population estimates.
Table A.4: Showing information needed for the Schnabel model to estimate population assuming a sampling period of one day with no same day recaptures included. Numbers using (CtMt)/Rt
0.00 0.00 0.00 0.00 0.00 134.40 248.50 285.20 299.17 328.33 338.95 374.18 499.36 437.20 448.51 500.38 637.02 756.32 755.40 852.88 949.32 934.10 975.16 945.27 1014.11 1147.56 1259.88 1354.84 1427.33 1611.83 1748.81 1789.45 1873.59 1923.81 1962.72 1961.12
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
0 0 1 0 28 23 28 12 9 15 15 16 46 8 24 26 48 38 16 30 65 45 70 54 38 64 62 58 55 101 100 91 109 71 41 7
0 0 0 0 1 4 3 2 2 3 4 3 6 7 8 5 5 4 6 3 15 19 24 29 7 9 12 13 15 19 26 37 39 27 14 4
0 0 1 0 27 19 25 10 7 12 11 13 40 1 16 21 43 34 10 27 50 26 46 25 31 55 50 45 40 82 74 54 70 44 27 3
0 0 0 1 1 28 47 72 82 89 101 112 125 165 166 182 203 246 280 290 317 367 393 439 464 495 550 600 645 685 767 841 895 965 1009 1036
0 0 0 0 28 644 1316 864 738 1335 1515 1792 5750 1320 3984 4732 9744 9348 4480 8700 20605 16515 27510 23706 17632 31680 34100 34800 35475 69185 76700 76531 97555 68515 41369 7252
Population estimate using all recaptures except same day catches.
Table A.5: Showing information needed for the Schnabel model to estimate population assuming a sampling period of one year Numbers using (CtMt)/Rt
0.00 0.00 0.00 0.00 0.00 307.50 895.00 663.50 565.33 579.75 550.18 560.29 741.33 586.56 565.09 626.97 813.40 951.91 927.28 1038.94 1283.30 1348.53 1547.04 1575.19 1716.86 1989.27 2149.12 2386.91 2540.40 2882.36 3179.07 3073.75 3216.63 3265.16 3326.72 3318.88
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
0 0 1 0 27 21 25 12 9 14 14 16 44 8 24 25 46 38 16 30 53 32 51 33 34 59 59 48 46 92 84 80 88 59 34 5
0 0 0 0 0 2 0 2 2 2 3 3 4 7 8 4 3 4 6 3 3 6 5 8 3 4 9 3 6 10 10 26 18 15 7 2
0 0 1 0 27 19 25 10 7 12 11 13 40 1 16 21 43 34 10 27 50 26 46 25 31 55 50 45 40 82 74 54 70 44 27 3
0 0 0 1 1 28 47 72 82 89 101 112 125 165 166 182 203 246 280 290 317 367 393 439 464 495 550 600 645 685 767 841 895 965 1009 1036
0 0 0 0 27 588 1175 864 738 1246 1414 1792 5500 1320 3984 4550 9338 9348 4480 8700 16801 11744 20043 14487 15776 29205 32450 28800 29670 63020 64428 67280 78760 56935 34306 5180
Population estimate using an assumed annual sampling period
Table A.6: Showing information needed for the Schnabel model to estimate population assuming a sampling period of 30 days
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
0 0 1 0 27 23 26 12 9 15 14 16 44 8 24 25 46 38 16 30 60 40 64 40 37 61 62 52 49 97 95 87 100 69 41 7
0 0 0 0 0 4 1 2 2 3 3 3 4 7 8 4 3 4 6 3 10 14 18 15 6 6 12 7 9 15 21 33 30 25 14 4
0 0 1 0 27 19 25 10 7 12 11 13 40 1 16 21 43 34 10 27 50 26 46 25 31 55 50 45 40 82 74 54 70 44 27 3
0 0 0 1 1 28 47 72 82 89 101 112 125 165 166 182 203 246 280 290 317 367 393 439 464 495 550 600 645 685 767 841 895 965 1009 1036
0 0 0 0 27 644 1222 864 738 1335 1414 1792 5500 1320 3984 4550 9338 9348 4480 8700 19020 14680 25152 17560 17168 30195 34100 31200 31605 66445 72865 73167 89500 66585 41369 7252
Numbers using (CtMt)/Rt
0.00 0.00 0.00 0.00 0.00 167.75 378.60 393.86 388.33 402.50 416.27 446.44 615.27 512.28 509.19 570.49 743.82 876.58 862.15 969.40 1108.60 1098.22 1152.61 1154.98 1240.30 1420.88 1544.43 1685.04 1791.79 2025.92 2185.51 2190.19 2284.24 2318.34 2348.86 2341.62
Population estimate using assumed 30 day sampling period
Table A.7: Showing information needed for the Schnabel model to estimate population for each area A, B and C assuming a 1 day sampling period to get a lower overestimate. (no mortality so all estimates are over estimates).
1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
72.16 97.626374 107.52137 110.85124 117.14173 135.00758 148.88387 155.36686 168.76796 174.215 191.14938 201.51341 212.53737 216.50842
8.8 19.441176 41.054054 47.775 58.431818 70.3125 58.928571 73.597826 74.173469 83.436364 96.85 127.21212 161.72993 173.51049 204.09459 220.68831 223.56129 4 9.5675676 28.5 44.857143 69.925926 84.835294 106.58511 118.52475 144.95556 149.8169 159.79874 166.21687 169.41667
8.8 19.441176 41.054054 47.775 58.431818 70.3125 58.928571 73.597826 74.173469 83.436364 96.85 127.21212 161.72993 173.51049 204.09459 220.68831 295.72129 101.62637 117.08894 139.35124 161.99888 204.9335 233.71917 261.95197 287.29271 319.17056 340.96628 361.31215 378.75423 385.92508
Appendix 7: Log sheet used for data collection
Figure A.8: Log sheet used to collect data for the project. Data collected was similar to data collected by the Glasgow Museum Tagging Programme, but with the addition of start and finish time.
Appendix 8: Questionnaires used to collect economic data
Figure A.11: Questionnaire aimed at anglers onboard charter vessels to collect economic data
Figure A.10: Questionnaire aimed at skippers of charter boat businesses to collect economic data
Appendix 9: Weight charts for D. batis
Table A.8: Chart to convert Length and wingspan (inches) into Weight (lbs) for males, accurate to 5%. Adapted from SAH, created by Bill Little.
Table A.9: Chart to convert Length and wingspan (inches) into Weight (lbs) for males, accurate to 5%. Adapted from SAH, created by Bill Little.