INTRODUCTION
The harbour porpoise (Phocoena phocoena) is one of the smallest cetaceans, and has a large geographical range in cold coastal waters of the northern hemisphere (Read, Reference Read, Ridgway and Harrison1999). In the north-eastern Atlantic and the North Sea, harbour porpoises are more or less continuously distributed, and the existence of sub-populations or of a continuous population with a significant isolation by distance is still under study (e.g. Rosel et al., Reference Rosel, Tiedemann and Walton1999; Tolley et al., Reference Tolley, Rosel, Walton, Bjørge and Øien1999; Andersen et al., Reference Andersen, Ruzzante, Walton, Berggren, Bjørge and Lockyer2001; Fontaine et al., Reference Fontaine, Baird, Piry, Ray, Tolley, Duke, Birkun, Ferreira, Jauniaux, Llavona, Öztürk, Öztürk, Ridoux, Rogan, Sequeira, Siebert, Vikingsson, Bouquegneau and Michaux2007).
Knowledge of the abundance and population structure of such species is essential for any conservation efforts (Tregenza et al., Reference Tregenza, Berrow, Hammond and Leaper1997; Hammond et al., Reference Hammond, Berggren, Benke, Borchers, Collet, Heide-Jorgensen, Heimlich, Hiby, Leopold and Øien2002). For some years, there has been growing concern about the size of harbour porpoise populations and their localization, and this has led to the formation of ASCOBANS (Agreement on the Conservation of Small Cetaceans in the Baltic and the North Seas). Some human activities are thought likely to have contributed to a decrease in the harbour porpoise population, especially in the south of the North Sea and management measures for porpoise conservation have been sought (ASCOBANS, 2008). Area specific analyses are required, as local specificities can be found. For instance, Tregenza et al. (Reference Tregenza, Berrow, Hammond and Leaper1997) estimated the impact of by-catch in set gillnets to be worrying for the Celtic Sea harbour porpoise population, and Northridge & Hammond (Reference Northridge and Hammond1999) evaluated the mortality caused by by-catch in the North Sea and West Scotland waters.
In 1994, harbour porpoises were shown to be strongly present in the northern part of the North Sea, whilst the Channel and the southern part of the North Sea had few sightings (Hammond et al., Reference Hammond, Berggren, Benke, Borchers, Collet, Heide-Jorgensen, Heimlich, Hiby, Leopold and Øien2002). Since then, work has shown that porpoise abundance has shifted markedly from the north to the south (Macleod et al., Reference Macleod, Scheidat and Hammond2006), and the Celtic Sea has become a region where harbour porpoises are commonly encountered. This new distribution has been confirmed by several local studies (e.g. Camphuysen, Reference Camphuysen2004; Kiska et al., Reference Kiska, Hassani and Pezeril2004, Reference Kiska, MacLeod, Van Canneyt, Walker and Ridoux2007; Thomsen et al., Reference Thomsen, Laczny and Piper2006).
A strong effort is being made in the French region of Brittany (north-western France) to monitor the local populations of marine mammals. Among these, the return of the harbour porpoise around the French Channel coast merits special attention. Since 1997 aerial survey, opportunistic sightings made by volunteer sea users, and stranding counts and analyses undertaken by a network of correspondents have been organized. These approaches have enabled us to monitor the local harbour porpoise population.
MATERIALS AND METHODS
Aerial survey and opportunistic sightings
In 2005, the LEMM (Laboratoire d'Etude des Mammifères Marins—Océanopolis, Brest, France) conducted an aerial survey along Brittany coasts (north-western France) according to standard line-transect methodology. Tracklines were surveyed following a saw-tooth pattern (Figure 1) designed for a high-winged twin engine aircraft (Partenavia P68C) flying at an altitude of 600 ft (182 m) at a speed of 100 knots. Bubble windows allowed the two lateral observers to see directly below the aircraft, while a third person noted the observation data. Encounter rates were calculated as (n/l) × 100 (n, number of encounters; l, total distance surveyed, in km).
Fig. 1. Distribution of harbour porpoise (Phocoena phocoena) sightings around the Brittany coastline (north-western France) from aerial survey (2005) and opportunistic sightings (2002–2007).
Since 2002, the collection of opportunistic sightings of marine mammals has been organized by the LEMM along Brittany coasts, in the framework of a dedicated regional scheme called ‘Observons la mer’. In addition to accredited observers, data were also recorded by different kinds of marine observers and users, such as ornithologists, fishermen, and sailors who were invited to collect data through the distribution of sightings reporting forms and posters in 1300 coastal locations around Brittany.
The LEMM regrouped and analysed all the data. Precise species were assigned only when sighting descriptions, and if available, pictures, were judged to be correct and sufficiently accurate. In this paper, only those reports of sightings of the harbour porpoise with a certain species identification collected between 2002 and 2007 are considered.
Stranding data
Since the beginning of the 1970s, the CRMM (Centre de Recherche sur les Mammifères Marins, La Rochelle, France) has coordinated a national marine mammal strandings recording programme. With the help of field correspondents, from organizations or volunteers, it has set out to collect data on marine mammal strandings originating from all over the French coastline.
Since 1995, the LEMM (Océanopolis, Brest, France) has coordinated this network at a regional scale, and collected data from Brittany coasts before sending them to the CRMM. The Brittany coasts have been divided into 18 sections extending to cover the whole of the coastline (Figure 2). In each of these areas, correspondents are trained in the analysis of stranded marine mammals: species identification, characteristic measurements and tissue sampling. The data collection effort followed a standard procedure in order to minimize possible variations between months or years. Species identification, sex determination and lengths of the animals were collected whenever possible. By-catch casualties were diagnosed following the criteria proposed by Kuiken (Reference Kuiken and Kuiken1994). This network allowed the LEMM to compile valid data on a large proportion of cetaceans stranded on the Brittany coasts.
Fig. 2. Distribution of harbour porpoise stranding events along the Brittany coastline (north-western France) between 1997 and 2007. Numbers and surfaces of the circles represent the numbers of stranded porpoises.
RESULTS
Aerial and opportunistic sightings of the harbour porpoise
The aerial survey of the coasts of the French region of Brittany was conducted during the summer of 2005, from 9–11 August. It covered 1578 km and enabled coverage of the whole of the Brittany coastline (Figure 1). Sightings of all cetaceans were recorded. A total of 24 schools of harbour porpoises (54 individuals) were observed, which represented 68.6% of the total cetacean schools sightings (34.6% of the individual sightings). The average porpoise group size was 2.25 individuals per group (SD = 1.92), and the encounter rate was 1.5 sightings per 100 km throughout the area studied. All the sightings were concentrated in waters around the western part of Brittany, and particularly on the north-western coast. This area corresponded to the shallow waters of the south-west waters of the western English Channel. In this area (377 km of aerial survey), the encounter rate peaked at 5.8 sightings per 100 km.
Between 2002 and 2007, 821 opportunistic cetacean sightings were reported, representing 11,419 individual animals. Among those, certain species identification was assigned to 222 reports, which included 3225 individuals in total. Harbour porpoise sightings were relatively rare, and represented only 9 reports (4% of the 222 cetacean reports) and 44 individuals (only 1.4% of the individuals sighted). For the most part, these observations were made in the same area where the aerial survey recorded harbour porpoise sightings, and during the same period (August), although some sightings were also localized in the south of Brittany (Figure 1).
Counting and analysis of stranding events
Between 1997 and 2007, the stranding data collection effort followed a standardized procedure, with a total of 135 harbour porpoises stranded along the north-western coasts of France being recorded. The region sampled included all French Brittany coastal areas, and a higher frequency of stranding was clearly observed in the western part of the area (Figure 2). Sex data were available for 80 individuals: 45 were identified as females, 35 as males.
Figure 3 shows the inter-annual distribution of the number of harbour porpoise strandings. An increase was observed from 1997 to 2007 (Figure 3A). This time-dependent increase is significant, as confirmed by the product-moment correlation calculation (r = 0.72, df = 9, P < 0.05). The same observation was made when the proportions of porpoises, among all stranded cetaceans, were determined for each year (Figure 3B): a clear increase was seen (r = 0.71, df = 9, P < 0.05). Conversely, stranding of cetaceans in general did not increase during the same period, and no correlation was observed between time and the numbers of stranded cetaceans recorded (r =−0.21, df = 9, P > 0.05). The classification of stranded porpoises on the basis of length more or less reflects their maturity status (Read, Reference Read, Ridgway and Harrison1999). The majority, almost three-quarters of all stranded porpoises, were longer than 130 cm, and can be classified as adults (Figure 3C). During the 11-year monitoring period, only six calves (smaller than 100 cm) were found stranded.
Fig. 3. Inter-annual distribution of harbour porpoise stranding events along the Brittany coastline (north-western France), and length distribution of stranded individuals. (A) Numbers of stranded porpoises from 1997 to 2007; (B) proportion of porpoises among all the stranded cetaceans, from 1997 to 2007; (C) length distribution of the porpoises stranded between 1997 and 2007.
Cetacean strandings, and especially those of harbour porpoises, occurred over the whole year. However, porpoise strandings were not uniformly distributed over the year (Figure 4A). More stranding events were recorded from the period extending from the start of winter to the beginning of spring (December–April). In fact, the distribution of porpoise strandings differed significantly from a 12-part uniform distribution (χ2(11) = 46.95, P ≪ 0.01).
Fig. 4. Intra-annual distribution of porpoise stranding events along the Brittany coastline (north-west of France) between 1997 and 2007. (A) Cumulated numbers of stranded porpoises for each month of the year. Greyed parts represent those porpoises for which a prior by-catch has been proved; (B) monthly proportions of porpoises among all the stranded cetaceans.
Figure 4B shows that there was an increase in the number of porpoises as a proportion of all cetacean stranding events in the period extending from September to January, as compared with the rest of the year: 12.7% of stranding events concerned porpoises between September and January, and only 6.7% during the remainder of the year. These proportions differed significantly (Z = 3.81, P ≪ 0.01).
For some stranded individuals, a by-catch prior to stranding can be proved by typical criteria (Figure 4A). Here, the winter months from December to February revealed higher proportions of proved by-catch (32.7% of porpoises stranding events) than during other months of the year (12.8%), and these differences appeared significant (χ2(1) = 6.50, P = 0.0108).
DISCUSSION
From the 1940s onwards, harbour porpoises have become scarce in the southern part of the North Sea, and around a very large zone of the European mainland coast, including the whole of Brittany (Duguy, Reference Duguy1977; Reijnders, Reference Reijnders1992). Recently, Scans-I and Scans-II (Hammond et al., Reference Hammond, Berggren, Benke, Borchers, Collet, Heide-Jorgensen, Heimlich, Hiby, Leopold and Øien2002; Macleod et al., Reference Macleod, Scheidat and Hammond2006) have detected a large-scale north to south shift of the harbour porpoise in the North Sea, and stable populations appear to be established again, as can be judged by sightings and stranding monitoring (e.g. Camphuysen, Reference Camphuysen2004; Siebert et al., Reference Siebert, Gilles, Lucke, Ludwig, Benke, Kock and Scheidat2006; Thomsen et al., Reference Thomsen, Laczny and Piper2006). Around the Brittany coast, it has been noted by some sea users that harbour porpoises can now be observed once more (Kiska et al., Reference Kiska, Hassani and Pezeril2004). In this study, we have tried to describe the occurrence of the harbour porpoise around the Brittany coast, and to confirm its comeback using different sources of data collected recently or currently in Brittany.
Occurrence of the harbour porpoise around the Brittany coastline
In Brittany, the regional scheme of opportunistic sightings, called ‘Observons la mer’, has been active since 2002, but when applied to harbour porpoises, it was not very productive. Between 2002 and 2007, 222 cetacean sighting forms with certain species identification have been filled in, and only 9 of them concerned harbour porpoises. In fact, the harbour porpoise is a very inconspicuous species, its behaviour is undemonstrative, and, moreover, it is much less well-known than other more popular species strongly represented around Brittany coasts, such as the common dolphin (Delphinus delphis) and the bottlenose dolphin (Tursiops truncatus). Furthermore, these observations were made only by accredited naturalists, and were therefore relatively rare events. As already pointed out (e.g. Siebert et al., Reference Siebert, Gilles, Lucke, Ludwig, Benke, Kock and Scheidat2006), opportunistic sightings are therefore not very useful when it comes to determining porpoise distribution patterns. We can nevertheless note that six opportunistic sightings out of the nine concerning harbour porpoises were made during the month of August, which perhaps represented a favourable month, in particular as regards the weather conditions for porpoise sighting along the Brittany coasts.
Standard line transect aerial surveys have been used on many different occasions to investigate porpoises and have proven to be efficient (e.g. Hammond et al., Reference Hammond, Berggren, Benke, Borchers, Collet, Heide-Jorgensen, Heimlich, Hiby, Leopold and Øien2002; Scheidat et al., Reference Scheidat, Kock and Siebert2004; Siebert et al., Reference Siebert, Gilles, Lucke, Ludwig, Benke, Kock and Scheidat2006; Thomsen et al., Reference Thomsen, Laczny and Piper2006). Harbour porpoises are believed to spend nearly half of their life in the surface metre (i.e. from 0–1 m depth) of the water column (Westgate et al., Reference Westgate, Read, Berggren, Koopman and Gaskin1995), which clearly represented a characteristic consistent with an efficient aerial observation. In fact, aerial survey allowed us to count a much more widespread distribution of harbour porpoises around the Brittany coast, as this species represented the majority of all the cetacean schools sightings (68.6%).
Encounter rates were of the order of 1.5 per 100 km along the whole of the Brittany coastline, which is consistent with results obtained before 2002 in the same area and at the same time of year using a ferry-based survey in the English Channel and the Bay of Biscay (Kiska et al., Reference Kiska, MacLeod, Van Canneyt, Walker and Ridoux2007). Almost all the harbour porpoise sightings we detected were restricted to shallow waters in the south-west of the western English Channel. Harbour porpoises generally formed groups of a few individuals (Kiska et al., Reference Kiska, Hassani and Pezeril2004; Siebert et al., Reference Siebert, Gilles, Lucke, Ludwig, Benke, Kock and Scheidat2006; Thomsen et al., Reference Thomsen, Laczny and Piper2006; Kiska et al., Reference Kiska, MacLeod, Van Canneyt, Walker and Ridoux2007). Here again, with good consistency, we determined an average of 2.25 individuals per group with a high SD (SD = 1.92).
The aerial survey was conducted during the month of August 2005 only. It would be useful to repeat this type of monitoring at other periods in order to look at the seasonal distribution. In fact, harbour porpoise sightings have been shown to vary markedly all year round, with low occurrence months that can alternate with high occurrence months in the southern North Sea (e.g. Siebert et al., Reference Siebert, Gilles, Lucke, Ludwig, Benke, Kock and Scheidat2006, Thomsen et al., Reference Thomsen, Laczny and Piper2006). Finally, because of the low number of sightings made during this aerial survey, density estimation was not calculated (Buckland et al., Reference Buckland, Anderson, Burnham, Laake, Borchers and Thomas2001).
For its part, the monitoring of strandings was conducted continuously over ten years. A total of 135 stranded harbour porpoises was recorded and analysed. Strandings occurred around the whole of the Brittany coastline, and in every year since 1997. Autumn and winter months seemed to lead to an increase of strandings that could be explained by weather conditions. But harbour porpoises were distinguished from the other cetaceans as they were found more frequently—in proportion among all the cetaceans—stranded between September and January. This monthly distribution of strandings appeared to be very different from the summer peaks detected, for instance, in the Baltic and the North Sea (Hasselmeier et al., Reference Hasselmeier, Abt, Adelung and Siebert2004; Siebert et al., Reference Siebert, Gilles, Lucke, Ludwig, Benke, Kock and Scheidat2006). Seasonal movement of the species that could approach the coasts and variations in the total distribution of the different cetacean species could be at the origin of this characteristic.
By-catch in fishing nets can lead to the death of harbour porpoises (e.g. Tregenza et al., Reference Tregenza, Berrow, Hammond and Leaper1997; Northridge & Hammond, Reference Northridge and Hammond1999; Siebert et al., Reference Siebert, Gilles, Lucke, Ludwig, Benke, Kock and Scheidat2006). In the Celtic Sea, the impact of fisheries using set gillnets on porpoise populations has been shown to be not negligible (Tregenza et al., Reference Tregenza, Berrow, Hammond and Leaper1997). During winter months, more than 30% of the porpoises stranded around the coasts of Brittany exhibited signs typically resulting from a by-catch incident (Kuiken, Reference Kuiken and Kuiken1994) prior to, and most probably the reason for, the stranding.
The local specificities highlighted by our study—an increased frequency of stranding during autumn and winter months and a strong impact of by-catch on stranding numbers during winter months—need further study. Harbour porpoises are known to be metabolically very active animals, and require significant daily food intakes (Evans, Reference Evans1987). The availability of foods, as well as interactions with other species in competing for the same food source, will also need to be studied during these periods of the year.
Harbour porpoise comeback in Brittany
Our study strongly confirms the comeback of the harbour porpoise around the Brittany coast. In fact, a yearly increase in the numbers of porpoise strandings has been observed, and the significance of this increase was reinforced by finding that the proportion of porpoises among all the cetaceans increased similarly. The standardized organization of the stranding data collection strengthens the validity of our analysis, by avoiding monthly or yearly variations in effort. Thus, if strandings are more or less representative of a given population for marine mammals, this stranding increase also reflects the comeback of harbour porpoises along the Brittany coast. In fact, stranding data represent a first step in detecting variations in the cetacean population abundance (Evans & Hammond, Reference Evans and Hammond2004). The harbour porpoise is known to be a shelf-water inhabitant, and is often observed in shallow waters. The French coasts of the south-west of the western English Channel offers these types of conditions, and a stable population of harbour porpoise could be established again. But a feature of our analysis is the almost total absence of stranded calves, strongly differing from other studies (e.g. Hasselmeier et al., 2004; Siebert et al., Reference Siebert, Gilles, Lucke, Ludwig, Benke, Kock and Scheidat2006), and this raises in particular questions about breeding sites.
A new aerial survey has now to be planned, covering the whole of the year, and with the major aims of calculating harbour porpoise densities, detecting potential seasonal movements, especially those concerning the average distances from the coasts, and searching for mother–calf pairs. A genetic analysis of the stranded individuals has also been planned for some years, and the samples are in the course of being studied (Hassani et al., Reference Hassani, Kerviel, Le Borgne and Jung2008). Comparison with other porpoise populations is in progress, in order to look for the possible existence of local sub-populations.
These long term monitoring, diet and genetic studies are required for a better understanding of the local porpoise population, and for the effective implementation of a conservation plan.
ACKNOWLEDGEMENTS
We are grateful to all members of the Brittany Stranding Network for their help in collecting data and samples. Special thanks are extended to all the members of the Océanopolis marine mammals team, to Sandra Fuchs, and Vincent Kerviel. We are also indebted to Gilles Couix, who drew the map of the harbour porpoise strandings along the Brittany coastline, and to two anonymous referees whose comments significantly helped to improve the manuscript.
