Seabirds and Cetaceans Branch, Joint Nature Conservation Committee, 17 Rubislaw Terrace, .... in marine habitats throughout the year were given the max-.
IBIS 137: S147-Sl52
A method of assessing seabird vulnerability to surface pollutants J. M. WILLIAMS, M. L. TASKER*, I. C. CARTER & A. WEBB Seabirds and Cetaceans Branch, Joint Nature Conservation Committee, 17 Rubislaw Terrace, Aberdeen, Scotland ABl I X E , UK
Substantial information on the offshore distribution of seabirds exists for the North Sea and nearby areas. This information is extensive but is not easily applicable to management problems, such as oil pollution incidents. A quantitative oil vulnerability index has been developed, based on four easily scored factors, to assess the vulnerability of seabird species to surface pollution. The index has been applied to calculate area vulnerability scores for 15’ latitude x 3 0 longitude rectangles in the North Sea by combining the species vulnerability scores with information on seabird densities. Maps of vulnerable concentrations of seabirds for each month of the year have been published with interpretative text. These maps can be used as a management tool, for example, in assessing potential risk or in the event of an oil pollution incident. Seabirds face a range of threats, including habitat destruction, effects of toxic chemicals, decliningfish stocks and surface pollution. Although vegetable oils and some chemicals, such as nonylphenol and dodecylphenol, have resulted in some seabird deaths (Zoun & Boshuizen 1992). mineral oils pose the greatest hazard among the surface pollutants (Dahlmann et al. in press). Damage to seabird populations from an oil spill is difficult to demonstrate, but oil pollution kills large numbers of birds every year (Camphuysen 1989).and the risk remains that future spills could cause major damage. Offshore surveys of seabirds in the North Sea have been conducted,using standard methods (Tasker et al. 1984, F’ihl & Frikke 1992). by teams from several nations since 1980 (e.g. Tasker et al. 1987. Durinck et at. 1993). Data from all teams using these standard methods have been combined to form the European Seabirds at Sea database, enabling bird densities (birds/km2)to be determined for areas of sea. The database presently holds in excess of a million records. giving information on sightings of over 12.5 million birds. This number of records is too large for easy manipulation or interpretation by non-specialists. Therefore, an oil vulnerability index (OVI) was developed to assess and map the vulnerability of seabirds in the North Sea. This paper describes the development of the OVI and some of the background to the production of an atlas of bird vulnerability within the North Sea (Carter et al. 1993). A vulnerability index to assess the threat to birds from oil pollution was first proposed by King and !?anger (1979) for maritime birds in the northeast Pacific Ocean. Their index comprised 20 factors affecting bird survival under the headings of range, population, habits, mortality and annual exposure (to the marine environment).Each factor was given
* Author for correspondence.
a score of 0.1. 3 or 5 by an expert group, and the scores were summed to give a final index value for each species. A similar but slightlymore sophisticated index was proposed by Camphuysen (1989)for the North Sea. Speich et al. (1991) scored 14 elements to form three components of a “bird oil index” in North America. Their three componentswere vulnerability as determined by bird behaviour, vulnerability as determined by population characteristicsand sigmficance of the region to total North American population. Tasker and Pienkowski (1987) and Tasker et al. (1990) assessed the vulnerability of species to oil pollution on the basis of time spent on the water and the importance of U.K. waters to each species. On the basis of these assessments, seabirds were assigned to three categories of vulnerability.King and Sanger (1979) commented that there was a lack of information to score accurately many of the factors that they used, and this remains true today. We considered that, for an index to be a reliable method of assessing species vulnerability to surface pollution, there should be information available to score all the factors used.
METHODS Assessment of factors The OW used to assess the vulnerability of each species to surface pollutants in the North Sea is made up of the following four factors: (a) the proportion which was oiled of each species found dead (or moribund)on the shorelineand proportion of time spent on the surface of the sea by that species: (b) the size of the biogeographicalpopulation of the species: (c) the potential rate of recovery following a reduction in numbers for each species; (d) the reliance on the marine environment by each species.
S147
S148
J . M. W I L L I A M S E T A L .
Table 1. Proportion of each species that was oiled of those found dead (or moribund) on the shoreline and ratio of time spent on the surface of the sea compared with in flight by that species, used to score factor a
Yo birds oiled ~
Score
Ratio of birds on water to birds flying
Score
~~~
81-100 61-80 41-60 21-40 0-20
2.5 2.0 1.5 1.o 0.5
>7 5-7 3-5 1-3 0-1
2.5 2.0 1.5 1.0 0.5
Factor a: Proportion of each species that was oiled of those found dead (or moribund) on the shoreline and proportion of time spent on the surJace ofthe sea by that species
Factor a was included to indicate the risk of an individual seabird in a population being killed by chronic oil pollution. It was scored using a combination of information from surveys of dead (and moribund)beached birds around the North Sea and an analysis of the European Seabirds at Sea database. Beached bird survey results (Stowe 1982,Camphuysen 1989. Danielsen et al. 1990) provided information on the proportion of birds of each species that was oiled amongst those found dead or moribund on the shoreline. The European Seabirds at Sea database was analysed to determine the ratio of birds of each species recorded on the surface of the sea to birds flying over the sea during daylight. For each species, the beached bird survey results and the analysis of the European Seabirds at Sea database were graded into broad categories and each was given a score between 0.5 and 2.5 as shown in Table 1. These were added together to give a final score for factor a of between one and five. Species which are often found oiled by beached bird surveys and spend a large proportion of their time in contact with the sea’s surface scored most highly for this factor.
Table 2. Biogeographic population estimates used to score factor b. The geographic limits for each population vary. See text references for further details
Estimated biogeographical population (pairs) 1-50.000 50.00(1-150.000 150.000-400.000 4OO.OOO-1,000.000
1,OOO.OOO+
IBIS
137
Table 3. Scoresforpotential reproductiveoutput used to score Factor c. Eachfactor (i, ii, iii) is scored separatelyfrom 1 to 5. These scores are then added together and given a factor c score as shown in the second part of the table
Mean clutch size (i) 1 2 3 4-5 6+
Maximum Ageat clutch first breeding size (iii) (il) 1 2-3 4-5 6-7 8+
6+ 5
4 3 1-2
Score (i 5 4 3 2 1
Sum
Factor
12-15 9-11 7-8 5-6 54
5 4 3 2 1
+ ii + iii) c score
Factor b: Biogeographical population
Species with small populations were considered to be more at risk from pollution incidents than were those with large populations. The most up-to-date estimates of the whole biogeographical population of each species or sub-species occurring in the North Sea (Cramp& Simmons 1977,1983, Cramp 1985. Laursen 1989, Lloyd et al. 1991)were used to score factor b (Table 2).
Factor c: Potential rate of recovery following a reduction in numbers Species with the potential to produce many young per year were considered to be able to recover more rapidly and hence less at risk than species that produce fewer young per year. Factor c was scored using information on clutch size (average clutch size and maximum clutch size) and age at first breeding (Cramp & Simmons 1977.1983, Cramp 1985). Maximum clutch size was taken as the maximum number of eggs normally laid by a female in one calendar year. Exceptional clutch sizes recorded on a few occasions were not included. Age at first breeding was taken as the mean age from published information. Each of these factors was
Table 4. Values used to score the reliance on the marine environment (factor d) of each seabird species % numbers in North Sea reliant
Factor b score
on marine environment
5 4 3 2 1
91-100 81-90 61-80 41-60 1-40
Factor d score
1994
S E A B I R D V U L N E R A B I L I T Y TO S U R F A C E P O L L U T A N T S
scored between one and five as shown in Table 3. The scores were summed to give a final score, between 3 and 15. used to score factor c according to the ranges given in Table 3. Species such as the Fulmar Fulmarus glacialis, Gannet Morus bassanus and Guillemot Uria aalge that lay only one egg each year and do not breed until they are several years old have the highest score for this factor. At the other end of the scale are the seaducks: this group tends to have large clutches and usually commences breeding at 2 or 3 years of age.
S149
Sea (Table 5). This is due mainly to the amount of time these species spend in contact with the water’s surface plus their low reproductive rate which will lead to a slow recovery from a decrease in numbers. The score for auks is lower than might perhaps be expected due to (for most species) a moderate to large biogeographic population size. Cormorants are moderately vulnerable, for while they are dependant on the sea. they have relatively fast reproductive rates. The low values for most seaducks and gulls reflect their large populations and potential for rapid recovery from losses.
Factor d: Reliance on the marine environment Some seabirds use only offshore waters for feeding and resting, while others use both offshore and onshore habitats. Those that spend more time at sea were considered to be at greater risk than were those that also use land. Factor d was assessed using habitat information in Cramp (1985) and Cramp and Simmons (1977, 1983). The proportion of the numbers of birds in the North Sea reliant on the marine environment was estimated within broad bands for each season. The season during which the highest proportion of the population of each species is dependant on the marine environment was then used to score factor d (Table 4). Species that come ashore only to breed or feed exclusively in marine habitats throughout the year were given the maximum score for this factor, for example Manx Shearwater Puffinuspuffinus,Common Scoter Melanitta nigra and Arctic Tern Sterna paradisaea. Species where many individuals are not reliant on the marine environment at any time of year were given a lower score, for example Great Crested Grebe Podiceps cristatus and Black-headed Gull Larus ridibundus. A large proportion of these species would be expected to survive a marine oil pollution incident even if marine habitats in the North Sea were no longer available to them. The OVl formula For each species, the four factors were scored between one (low vulnerability) and five (high vulnerability),and the following formula was used to give an OVI score: OVI = 2a
+ 2b + c + d.
Factors a and b were given twice the weighting of factors c and d because we considered a and b the most important. The weighting is arbitrary but explicit and could be changed for future applicationsof the methodologyor if a more quantitative method for assigning weightings were devised. For each species, the scores assigned for each factor have been assessed on the basis of the best available information. The quality and quantity of information currently available vary between species, and it is possible to revise scores in the future or if the methodology is applied to different areas. The maximum OVI score is 30. The OVI values reveal that divers. grebes, skuas and auks are the most vulnerable groups of marine birds in the North
Using the index The OVI scores for individual species serve only to rank the relative vulnerability of species (Table 5) and are of no use on their own in assessing the vulnerability of offshore areas. No allowance is made within the OVI score for variable use of an area through the year. This requires knowledge of the relative importance of the area for each species in each month and the total number of species using the area. North Sea bird records held on the European Seabirds at Sea database provided this information. A density was calculated for each species listed in Table 5 for each 15 minutes of latitude by 3 0 minutes of longitude rectangle for which data were available. In order to assess total vulnerability of offshore areas, the density values were then combined with the species OVI scores using the formula area vulnerability score = ZSm, In(p + l)=OVI, where p is the density calculated for a species in the area and OVI the oil vulnerability index score for that species. The area vulnerability is thus a combination of the numbers of each species present in each area and their OVI scores. Natural logarithmswere used to transform each density into an order of magnitude. This smoothes out variations in the number of birds seen in a survey but still highlights largescale variations. One was added to each density to avoid negative logarithm values. The value for each rectangle was then placed into one of four categories of vulnerability (very high, high, moderate and low) by dividing the range of values into four equalsized groups. A map for each month of the year was then plotted. For wider usage in the North Sea atlas, it was necessary to include further information on birds vulnerable to oil pollution but not assessed using the techniques described above. These birds include shorebirds, some concentrations of inshore waterfowl and features such as night-time gull roosts (see Carter et al. 1993 for detailed sources). This information was also plotted on the maps. For ease of interpretation, the resulting plots were then smoothed by hand. The halised maps (e.g. Fig. 1) were published with interpretative text (Carter et al. 1993). On these finalised maps, areas holding high densities of the most vulnerable species are shown as dark shading, with progressively lighter shading indicating areas of lesser vul-
S150
J. M. W I L L I A M S ET AL.
IBIS
137
Table 5. Oil vulnerability index (OW) scoreslot- seabird species regularly occurring in the North Sea
Factor Species Red-throated Diver Gavia stellata Black-throated Diver Gavia arctica Great Northern Diver Gavia immer Black Guillemot Cepphus grylle Red-necked Grebe Podiceps grisegena Great Skua Catharacta skua Shag Phalacrocorax aristotelis Arctic Skua Stercorarius parasiticus Little Gull Larus minutus Razorbill Alca torda Great Crested Grebe Podiceps cristatus Manx Shearwater Puffinus puffinus Gannet Moms bassanus Guillemot Uria aalge Little Auk Alle alle Velvet Scoter Melanitta fusca Red-breasted Merganser Mergus serrator Great Black-backed Gull Larus marinus Puffin Fratercula arctica Cormorant Phalacrocorax carbo Scaup Aythya marila Sandwich Tern Sterna sandvicensis Common Tern Sterna hirundo Sooty Shearwater Puffinus griseus Common Scoter Melanitta nigra Lesser Black-backed Gull h r u s fuscus Little Tern Sterna albifrons Fulmar Fulmarus glacialis Storm Petrel Hydrobates pelagicus Long-tailed Duck Clangula hyemalis Kittiwake Rissa tridactyla Eider Somateria mollissima Goldeneye Bucephala clangula Arctic Tern Sterna paradisaea Herring Gull Larus argentatus Common Gull Larus canus Black-headed Gull Lams ridibundus
a
b
C
d
OVI score
5 5 5 5 4.5 3
5 5 5 5 5 5 4 5 5 2 5 3 3 1 1 3
4 4 4 4 2 4 3 4 3 5 2 5 5 5 5 1 1
5 5 5 5 5 5 5 5 5 5 3 5 5 5 5 5 5 4 5 4 5 5 5 5 5 3 5 5 5 5 5 5 3 5 3 2 2
29 29 29 29 26 25 24 24 24 24 23 23 22 22 22 21 21 21 21 20 20 20 20
4 2.5 3 5 4 3.5 3 5 5 4.5 3.5 2.5 4.5 2.5 3 1.5 1.5 3.5 4.5 3 1 3 1 4.5 3 4 3 1.5 2 2 2
nerability.Unsurveyedareas are shown in the lightest shade. In addition, symbols have been used to indicate the main type(s) of birds in concentrations. The size of the symbol indicates the importance of the concentration.
DISCUSSION King and Sanger (1979) considered that an index to assess species vulnerability would become ‘a useful management tool with all sorts of possible applications’.The OVI has been designed for use as an aid in assessing the vulnerability of concentrations of birds to surface pollutants, especially oil. The method has been applied to the North Sea and can easily be used for other areas.
4
4 1 4 4
4 4 1 2 3 4 1 3 1 1 1 3 2 2 2 1
4 5 3 1 4 4 5 1 4 4 5 5 1 4 1 1 4 4 3 3
19
19 19 19 18 18 17 17 16 16 16 15 13 11
We consider the technique useful in making the assessment of species and area vulnerability more quantitative and repeatable. This enhances the quality of conservation advicethat can be given,either in consultationsor in reactive situations following a pollution incident. In the event of a pollution incident, the relevant monthly map can be consulted to assess the importance to birds of the affected and adjacent areas. The style and speed of response to the pollution incident can be tailored to take account of the bird resource at risk. The atlas can also be used for contingency planning, for potential pollution incidents and in planning offshore operations that increase the risk of oil pollution, such as exploration drilling or the routeing of tankers. It is important to accept the limitations of the mapping process. The maps have been prepared from data collected
1994
s151
SEABIRD VULNERABILITY TO SURFACE POLLUTANTS
62 "N
61
60
59
58
57
56
55
54
53
52
51 I
5
4
3
2
1
0
1
2
3
4
"W
5
6
~
1
i
7
8
9
1 10
)
I
11
12
! 1 1 3 1 4
I
"E
Figure 1. A map of the North Sea area showing areas of varying vulnerability to surface pollution for seabirds and shorebirds (shading) and regions of concentration of birds indicated by symbols (after Carter et al. 1993).
over a 13-year period and can therefore show only average levels of abundance and hence vulnerability over that period. Seabirds are highly mobile and can respond to shortterm changes in environmental conditions and food availability. To obtain more accurate and up-to-date information for an area affected by a spill, the best option may be to carry out a rapid survey using standard methods. The OVI methodology could then be used to assess objectively the vulnerability of bird concentrations in the area surveyed. The indices proposed by King and Sanger (1979). Camphuysen (1989) and Speich et al. (1991) were based on up to 20 factors covering a wide range of the biological features of seabirds. However, they were scored mainly by ranking species rather than by assessing absolute values. Their indices can therefore be viewed only as semi-quantitativemeasures. Although the index presented here is based on only four factors, it has been possible to score each species objectively for each factor, and the combination of the species scores with density data has proved a valuable mechanism for assessing the relative vulnerability of areas. The index could
be improved if more factors could be assessed, but there is a lack of basic information for many species. As Wiens et al. (1984) noted, the sort of information which might be used to build models of the response of birds to marine pollution (for example, age-specific fecundity and survivorship: the probability of death on encountering an oil spill: the changes in resource availability following an oil spill: the proportion of the biogeographic population using an area) is difficult to collect or assess and is not available at present for most species. By combining the species vulnerability scores with densities in areas of the marine environment, the area vulnerability calculation attempts to balance the impact of the death of large numbers of a species in a small area against fewer over a larger area. It is difficult to decide which, if either, of these should be viewed as the more important, but it is certainly true that at some times of year several species are very vulnerable in small areas, for example the huge flocks of moulting seaduck in the western Kattegat. The index takes no account of the distribution or range of each
S152
J . M. W I L L I A M S ET A L .
species, and thus, though only a few birds might be killed by a small incident, these could form the whole or a large part of numbers locally. It is difficult to approach this question of scale, but the index was designed to be used to assess the importance of relatively large areas and considers the loss of individuals against the size of the total population of a species. The data which forms the European Seabirds at Sea database was collected by teams based in Belgium, Denmark, Germany, the Netherlands, Norway, Sweden and the United Kingdom. National datasets have been gathered with funding from a great many sources, both in the oil industry and from national governments. The U.K. Seabirds at Sea Team is currently sponsored by the Departments of Transport and of Trade and Industry, British Petroleum, Shell, Esso. Elf Enterprise, British Gas and Amoco. We thank all sponsors and those who gave help in kind for their contributions. The collation of national datasets to form the European Seabirds at Sea database and the production of the vulnerability atlas described in this paper were funded by the United Kingdom Departmentofthe Environment in a contract to the Joint Nature ConservationCommittee’s Seabirds at Sea Team. We thank the members of the database project coordination committee and members of the Seabirds and Cetaceans Branch (Per Andell, Henk Baptist, Kees Camphuysen. Jan Durinck. Arne Follestad. Stefan Garthe. Ommo Hiippop. Claude Jouris. Karsten Laursen, Mar& Leopold. Leif Nilsson, Stefan PihI, Henrik Skov, Carolyn Stone, Paul WaIsh) for their helpful comments and discussions. Tony Fox and an anonymous referee improvedthe manuscript with their comments.
REFERENCES Camphuysen, C.J. 1989. Beached Bud Surveys in the Netherlands 1915/1988. Seabird mortality in the Southern North Sea since the early days of oil pollution. Technisch Rapport Vogelbescherming 1. Amsterdam: Werkgroep Noordzee. Carter, I.C.. Williams. J.M., Webb, A. &Tasker. M.L. 1993. Seabird concentrations in the North Sea: A n atlas of vulnerability to surface pollutants. Aberdeen: Joint Nature Conservation Committee. Cramp, S. 1985. Handbook of the Birds of Europe, the Middle East and North Africa. The buds of the Western Palearctic, Vol. N. Oxford: Oxford University Press. Cramp, S. & Simmons. K.E.L. 1977. Handbook of the Buds of Europe. the Middle East and North Africa. The birds of the Western Palearctic, Vol. I. Oxford: Oxford University Press. Cramp, S. & Simmons. K.E.L. 1983. Handbook of the Buds of
IBIS
137
Europe,the Middle East and North Africa. The birds of the Western Palearctic. Vol. 111. Oxford Oxford University Press. DahImann. G.. Timm,D.. Averbeck. C., Camphuysen. C.J.. Skov. H. & Durinck. J. 1994. Oiled seabirds-Comparative investigations on oiled seabirds and oiled beaches in The Netherlands, Denmark and Germany (1990-1993). Mar. PoUut. B d . 28: 305-310. Danielsen. F., Skov. H.. Durinck, J. & Christensen. K. 1990. Seks &-s overvagning af dede havfugle. Dansk Ornit. Foren. Tidsskr. 8 4 8-9. Durinck. J., Skov, H. & Danielsen. F. 1993. Some factors determining seabird distribution in the North Sea. In Madsen. J. (ed.) Proc. 7th Nordic Congr. Omithol.. 1990: 9-17. King, J.G. & Sanger, G.A. 1979. Oil vulnerability index for marine oriented birds. In Bartonek. C.J. & Nettleship. D.N. (eds) Conservation of Marine Birds of Northern North America: 227-239. Wildlife Research Report No. 11. Washington. D.C.: U S . Department of the Interior. Fish and Wildlife Service. Laursen. K. 1989. Estimates of seaduck winter populations of the Western Palaearctic. Dan. Rev. Game Biol. 13: 1-22. Lloyd, C., Tasker. M.L. & Partridge, K. 1991. The Status ofseabirds in Britain and Ireland. London: T&AD Poyser. Pihl, S . & Frikke, J. 1992. Counting birds from aeroplane. In Komdeur, J., Bertelsen. J. & Cracknell. G. (eds) Manual for Aeroplane and Ship Surveysof Waterfowl and Seabirds:24-34. rwRBSpecial Publication No. 19. Slimbridge: International Waterfowl and Wetlands Research Bureau. Speich. S.M..Manuwal. D.A. & Wahl. T.R. 1991. The bidhabitat oil index-A habitat vulnerability index based on avian utilization. Wild. Soc. Bull. 19: 216-221. Stowe. T.J. 1982. Beached Bird Surveys and Surveillanceof Cliffbreeding Seabirds. Chief Scientist Directorate Report No. 366. Peterborough: Nature Conservancy Council. Tasker, M.L. & Pienkowski, M. 1987. Vulnerable Concentrations of Birds in the North Sea. Peterborough: Nature Conservancy Council. Tasker. M.L.. Jones, P.H.. Dixon, T.J. &Blake,B.F. 1984. Counting seabirds at sea from ships: A review of methods employed and a suggestion for a standardized approach. Auk 101: 567-577. Tasker. M.L.. Webb, A.. Harrison, N.M. & Pienkowski,M.W. 1990. Vulnerable Concentrations of Marine Birds West of Britain. Peterborough: Nature Conservancy Council. Tasker. M.L.. Webb, A.. Hall. A.J.. Pienkowski. M.W. & Langslow. D.R. 1987. Seabirds in the North Sea. Peterborough: Nature Conservancy Council. Wiens. J.A., Ford, R.G. & Heinemann. D. 1984. Information needs and priorities for assessing the sensitivity of marine birds to oil spills. Biol. Conserv. 25: 2 1 4 9 . Zoun, P.E.F. & Boshuizen. R.S. 1992. Gannets victim to spillage of lubricatingoil and dodecylphenolin the North Sea, winter 1990. Sula 6: 29-30.
