Barra Otter Survey 22-29 July 2000

This report deals with a survey carried out on Barra in July 2000;  its purpose was to obtain data on distribution and population density and to look at what factors influence the distribution of the otter population around the coastline. 

The Isle of Barra is one of the Outer Hebrides situated on the northwest seaboard of Scotland and its location is shown in Figure 1.  The island is about 18km x 11km  and it has a variety of scenery from the craggy hills such as Heaval (384m) to the broad grassy machair land.  The geology of the island is made up of Lewisian gneiss, which is  a strong metamorphic rock which occurs on all the Outer Hebrides.                                       

Acknowledgements

I would like to thank the following volunteers who took part in the survey:

Sara Calhim, Sherry Fuller, Alex Hardy, Helen Hardy, Margaret Hunter, The late Margaret Hutchinson, Sarah Jupp, Tracey King, Ruth Moffatt, John Noorani, Mary Taylor, Lisbeth Tuckey, Sylvia Sadler, Barbara Wild. 

METHODOLOGY 

For the purposes of this study the Barra coastline was divided into 140 sections of 500m in length measured at the High Water Mark.  Personal experience had shown that such sections are easily manageable for fieldwork and are long enough to observe otter activity from a set vantage point without losing a part of the section around a corner or blocked off by a hill.  For each section a data sheet was printed . A total of 72 coastal sections were examined, accounting for 51% of the total Barra coastline.  

In this report the following factors are examined 

     a)   Width of tidal zone

     b)   Height of point 25m from High Water Mark

     c)   Coastal type

d)      Inland vegetation

e)      Number of sprainting points

f)        Otter activity

g)      Number of holts

h)      Number of freshwater pools 

a) Width of Tidal Zone 

The width of the tidal zone was measured using the maximum width over the 500m section.  It was measured using the Ordnance Survey 1:25,000 map.  Since the maximum width is used in all sections it will give a standard benchmark for the width of the intertidal zone.

b) Height of point 25m from High Water Mark

The height of the point 25m from the High Water Mark was used as an indicator of the general slope of the shoreline and measured from the 1:25,000 maps.  The reading was taken from the centre of the 500m section at the relevant contour line nearest this 25m mark. 

c) Intertidal Make up

The intertidal coastal zone was divided into six shore types shown below. Excluding the rock outcrop all are related to grain size as shown in Table 1.  The shore type was estimated by eye and by looking at samples on the shore.  The shore type chosen for the 500m section was that which occupied the largest part of it (usually >50%). The maximum diameter of the boulder was measured using a pocket tape measure. 

Muddy

The term “mud” was used according to the definition given by Gore (1997), which describes it as a mix of silt and clay and defines “mud” as a substrate containing 15% sand, 45% silt and 40% clay.  Clay particles are defined as any particle less than 0.0625mm in diameter. 

Muddy shores require sheltered conditions to form, and in places vegetation is dominated by the red algae Phyllophora crispa, and in the sublittoral zone by Laminaria saccharina and Chorda filum. 

Sandy

Sand grains are by definition between 0.0625-2mm in diameter (Read, 1972).  The definition of sand usually means quartz sand, but other sands can be found:   other minerals and shell fragments can also be mixed with quartz to form a sand. 

Sandy coasts have poor vegetation communities and are too mobile to support algae. Isolated patches of Laminaria saccharina may be present with beds of Zostera marina in the littoral zone. 

Shingle

Pebble, gravel or shingle deposits range in grain size from 2-64 mm (Blatt et al, 1972). Shingle is a mixture of material accumulated on beaches or offshore bars.  The shingle shores make up one of the poorest maritime habitats.  Thick beds of Phymatolithon calcareum and Lithothamnium corallioides may occur on some shores and in places Laminaria saccharina may be present in the sublittoral zone. 

Boulder

Boulder shorelines were classified as those with individual particle sizes ranging from 64 mm upwards and consisted of eroded country rocks and glacial material. 

They were divided into two size groups as follows:

i.   > 20 cm

ii.  < 20 cm 

The littoral region is usually dominated by Ascophyllum nodosum and Fucus serratus on sheltered coasts and Fucus vesiculosus on more exposed areas.  In the sublittoral zone the encrusting algae Lithothamnium  corallioides may be common. 

Rock Outcrop

When the dominant coastal type was bare rock, this category was used to define the intertidal zone.  Rock outcrop could be steep eroded country rocks or gently-sloping wave-cut platform.  

The communities on the rocky coastlines generally consist of Ascophyllum nodosum and Fucus serratus in the littoral region with the sublittoral region dominated by Laminaria saccharina and Laminaria digitata on more sheltered coasts replaced by Laminaria hyperborea and Fucus vesiculosus on exposed coasts. 

One of the above categories was allocated for each 500m coastal section.  When more than one category was present in the zone, the dominant category in that 500m coastal zone was allocated.  The "boulder" category was further divided into boulders greater or less than 20cm 

d) Inland Vegetation

Inland vegetation was recorded 20m from High Water Mark and the categories used are shown below. 

Heather

The heather moors are dominated by Ling (Calluna vulgaris) although this is not the only heather species present:  Bell Heather (Erica cinerea) and Cross-Leaved Heath (Erica tetralix) occur frequently with Ling.  Heather grows on well-drained soils, but hollows may occur with grasses, sedges and rushes in these waterlogged areas. 

Scrub

Scrub habitat consisted of a mixture of plants including Willow (Salix sp), stunted Birch (Betula pendula), Bramble (Rubus sp) and Bracken (Pteridium aquilinum).  

Grassland

These are areas that are extensively grazed by sheep and cattle.  The pastures have been modified by cultivation and draining and the grasslands have communities of tall herbs associated with the production of hay.  

Native Wood

Native woods are scarce on Barra, and are restricted to gorges and small coastal areas out of the reach of grazing animals and fire.  The woods are dominated by Birch (Betula pubescens), and Hazel (Corylus avellana), with some Rowan (Sorbus aucuparia), Holly (Ilex aquifolium) and Oak (Quercus petraea). 

e) Numbers of Sprainting Points 

Sprainting points were defined as places where one or more spraints or piles of spraints were found.  In the coastal environment, spraints are found most frequently on rocky coasts, on well-marked sites at the mouths of rivers, on otter runs or at certain times of the year at the entrance to holts (Conroy and French, 1985, 1987).  The droppings add nitrogen to the surrounding area making sprainting points very obvious:  grassy areas tend to be very green with a prominent growth of nitrophylous grasses. 

f) Direct observations of otters  (with telescope and binoculars)

Observations were made during the survey in each 500m section using binoculars and a X30 telescope. 

g) Number of freshwater pools

A freshwater pool was defined as an area of standing water with a minimum width or length of 0.5m and minimum depth of 0.1m.  It has been found from personal observation that otters show no signs of utilising pools with dimensions less than this.  All freshwater pools, which met these criteria, were counted. 

h) Numbers of holts

Otter holts are very variable and for the purpose of this study they were divided into two distinct types. 

Holts: Defined as a tunnel system with signs of regular use by otters (Kruuk et al, 1989).  The evidence of use consisted of tracks, flattened grass, fresh spraints and the characteristic odour in the holt.  On Skye, I found holts consisted dominantly of rock piles within 50m of the shore so such holts were also included in the data for Barra.  If two holts were more than 10m apart they were considered separate. 

Lie-ups:  Otters use resting places throughout their home range and these are mostly hollows under rocks or within reed/rush beds.  They have a path leading to them and a sprainting point associated with them.  

For the purpose of analysis only the holts and not lie-ups were included in the count of holt numbers. 

Results 

The volunteers carried out a survey of 72 x 500m coastal sections accounting for 52% of the Barra coastline and representing 36 hours observation of the coastal sections.   

Environmental variables 

Grassland accounted for 93% of the vegetation adjacent to High Water Mark with heather moorland accounting for 7%.  Of the area examined rock outcrop dominated the intertidal make up accounting for 80.5%.   

Comparing the height 25 m from High Water Mark and the width of the intertidal zone with the east and west coast it is seen that the heights did not differ significantly, but wider intertidal zones occurred on the east coast.  

Otter Activity 

 The mean number of sprainting points was greater on the east side of the island than on the west, however a chi-squared test showed no statistically significant differences between the east coast and west coast with regard to otter activity.  (χ2 = 3.08, df =7, p>0.05).  

Otter sightings were scarce on Barra;  two otters were seen briefly at Ledaig and one of our party who stayed on for an extra day saw a mother and two cubs under the War Memorial on the way to Vatersay. 

In Shetland there was a good correlation between otter numbers and holt numbers. (Moorhouse, 1988, Kruuk et al, 1989), enabling the number of holts to be used as an index for the number of otters.  The same was true on Skye (Yoxon,1999).  Because no detailed work over a long period could be undertaken on Barra the Shetland and Skye relationships were used to estimate otter numbers. 

In  Shetland following intensive work over many years, the relationship was found to be: 

0.33 females per holt 

Total number of otters = 1.83 X number of resident females 

On Skye four study areas were monitored over four years and the relationship was found to be: 

0.52 females per holt 

Total number of otters = 1.64 X resident females 

On Barra 46 sections out of 92 were surveyed on the east coast and 26 out of 48 on the west coast and the holt counts were: 

East coast:  0.22 holts per section

West coast:  0.15 holts per section 

From this it can be estimated that the total number of holts for the whole Barra coastline was 27.4. 

Using this data for all sections to estimate the total Barra otter population gives a figure of 16.6 otters based on the Shetland relationship and 23.4 otters based on the Skye relationship. 

Diet Analysis using Spraints 

In the coastal environment, spraints are found most frequently on rocky coasts, on well-marked sites at the mouths of rivers, on otter runs or, at certain times of the year, at the entrance to holts (Conroy and French, 1985, 1987).

Analysis of spraints 

Only fresh spraints were collected for analysis and stored in labelled plastic canisters.  In the laboratory, mucus was removed using Co-op false teeth cleansing solution (Composition greatest first:  sodium chloride, sodium percarbonate, trisodium phosphate, magnesium carbonate, methol, peppermint oil and sodium lauryl sulphate).  The samples were soaked in the solution for 24 hours then washed and sieved using a 0.5mm sieve.  This method has been used by many workers including Erlinge (1967, 1968); Watson, (1978) and  Beja, 1995. 

The bulk of the prey remains consisted of fish bones with occasional bird feathers and small mammal bones.  The prey was identified by comparing the fish vertebrae, bird feathers and small mammal bones with a reference collection and also compared with the published keys of Watson (1978) and Conroy et al (1993).

No method of expressing the results of spraint analysis describes the diet of otters accurately (Wise et al, 1981;  Kyne et al, 1989;  Rowe-Rowe, 1977;  Jenkins et al, 1980;  Heggberget, 1993;  Carss, 1996);  however, the two most commonly used methods are described by  Conroy et al  (1993): 

a) Percentage Frequency of Occurrence, where all the remains of each species in a spraint are taken to represent only one specimen and its frequency is calculated as the number of times it appears in all spraints in the sample.  This is calculated as: 

Number of spraints containing a particular prey item     

__________________________________________      X 100

Total number of spraints in sample 

b) Relative Frequency of Occurrence, where all the bones are identified in every spraint, and the frequency of each species in that spraint determined.  This is calculated as: 

Number of occurrences of a prey in each spraint

___________________________________________               X 100

Total number of all prey items identified in the spraint 

These methods, however, have important limitations, which will be considered in the Discussion. 

In this report the results of spraint analysis have been analysed by Percentage Frequency of Occurrence.  

RESULTS 

Total spraint composition

20 spraints were analysed, yielding 69 individually identifiable remains of 13 prey types. 

The Percentage Frequency of Occurrence of prey remains shows that marine fish dominated the diet (89%), with the most common prey species being Five-Bearded Rockling (23%), Viviparous Blenny (20%), Saithe (8%), Flatfish (7%) and Butterfish (5%).  It is worth noting the importance of Saithe which constitutes 8% of the prey remains and is the only free-swimming species as the other four are small benthic fish.  Small mammals played a small part  (5%) and seabirds (1.5%). The results are seen in Figure 1. 

Variation in spraint composition along the East and West coasts 

 A Spearman Rank Coefficient shows that a highly significant correlation occurs between the two coasts (r_s = 0.61, df=19, p=0.005) and therefore no difference in the populations between these two coastlines can be attributed to diet.  Small variations did however occur with more Blenny being found on the East coast and more Rockling, Saithe, Sea Scorpion, Flatfish and Common Eel being found on the West coast. 

Comparisons with Other Studies

Studies of the diet of coastal otters have been undertaken in Shetland (Kruuk et al 1987; and Kruuk and Moorhouse, 1990), in Norway (Heggberget and Moseid, 1994), on Loch Broom (Mason and Macdonald, 1980), Mull (Watt, 1995) and Skye (Yoxon, 1999) and the data present some interesting comparisons with this study. 

Significant correlations occurred between the data for Barra and Skye (r_s = 0.64, df=19 p=0.003), with similar correlations between Loch Broom, Shetland and Mull. 

The important conclusion is that the relative occurrence of fish species in spraints from Barra and the other islands is constant with a dominance of five key prey species (Viviparous Blenny, Five-Bearded Rockling, Butterfish, Sea Scorpion, Flatfish and Saithe)   

DISCUSSION 

Based on the data, 27.4 major holts have been identified on the Barra coastline; this gives an estimate for the total number of otters of between 16 and 23 otters. The estimate for coastal otters on Barra is derived by working out a relationship between resident females and active holts and a relationship between resident females and other otters, using research from Shetland and Skye.  This estimate is based on the numbers present at a particular point in time plus births, minus deaths, plus immigrants minus emigrants.  Stubbs (1977) has shown that animals from permanent habitats like the vertebrates tend to show an under compensating or exactly compensating mortality, and this would clearly be relevant when considering the Barra otter population.  The estimate given here is however likely to be an underestimate because of otters living in inland freshwater systems.  From work done on Skye it would appear that on such islands otters using the freshwater areas also use the coast and so they would presumably be included in the population numbers given (Gentle, personal communication. 

The figure of between 16 and 23 is based on this survey of 52% of the coastline and the following assumptions: 

1.  52% of the Barra coastline is a fair representation of the entire Barra coastline.  The sampling of 52% is very high compared with the 35% achieved in studies on Shetland (Kruuk et al, 1989), (Jolly, 1969).   

2.  All otter holts were found during the survey.   Some holts may have been overlooked due to the extensive nature of the area surveyed and this would have underestimated holt numbers.  This underestimation could be expected to have occurred equally on all coastal sections.   

3.  The total number of otters in the study areas was estimated correctly.  We do not know if the same relationship exists between resident females and holts on Barra but we can assume this to be the case as it has already been the case on Shetland and Skye.  The Skye data may  be assumed to give a better figure as Lewisian gneiss is found on Skye and on Barra. 

4. The relationship between the sexes of the resident otters and the transient animals was the same in all areas as in the study areas.  This was relatively constant in the four study areas on Skye and the ratio of cubs to females was also constant, so this can also be assumed to be the case on Barra. 

Diet 

The diet of the otter obtained from spraint analysis showed that five key prey species made up 74% of the diet (Viviparous Blenny, Five-Bearded Rockling, Butterfish, Saithe and Flatfish).  Clearly these species are of critical importance for the survival of the otter around the coastline not just of Barra but most of Northern Scotland. 

Final Remarks 

It has been shown that otter numbers are low on Barra  (16-24), but still an important part of the Eurasian otter population in Europe. In view of the low numbers of otters found on Lewisian gneiss rock types the latest estimated number of otters of  1,200 for the Outer Hebrides should be reviewed  (Harris et al 1995).  

Otter populations are linked to the number of freshwater pools in a particular section.  Freshwater pools are used by otters to maintain the insulating properties of their fur by washing out the salt contamination (Kruuk and Balharry, 1990).  Otters, unlike most marine mammals, have a very thin adipose tissue layer (Pond and Mattacks, 1985), (Tarasoff, 1974), and rely instead on a thin layer of air trapped in their fur (Kruuk and Balharry, 1990).  Kruuk and Balharry (1990) also showed that the fur of an otter lost much of its thermal insulation after five soakings in salt water. 

The necessity for the Eurasian otter to visit freshwater pools frequently has been demonstrated from observations in Shetland (Kruuk and Balharry, 1990), and Skye (Yoxon, 1999) and also from other species of otter:  the Cape Clawless Otter (Aonyx capensis) in South Africa where Van der Zee (1981) observed them frequently visiting freshwater pools after foraging in the sea; and Dr Carlos Olavarria in Chile (personal communication) showed that the Marine Otter (Lutra felina) visits freshwater pools or streams after being immersed in salt water.  It has been well documented by Kruuk et al (1989) that there is a dependence on freshwater pools, which were strongly, correlated with the occurrence of otter holts along the Shetland coastline.  

It is also clear that geology plays an important part in the distribution of freshwater pools along the Isle of Skye coastline, with correlations between geology and otter utilisation. 

Lovett (1994) and Lovett et al (1997), also looked at freshwater pools on the Torridonian sandstone coastline on the Isle of Skye and showed that pool utilisation was related to three factors, namely pool depth, amount of short grass surrounding the pool and the percentage of pool substrate composed of flat rock. 

In work undertaken by the International Otter Survival Fund, Priestley (1996) found a positive correlation between salinity of the pool and the presence or absence of spraint.  Sprainting was concentrated within a very narrow salinity range and there is a clear relationship between sprainting and low salinity pools.  Although he did not find a critical salinity above which otters would disregard a pool for bathing, he clearly showed that the otter prefers to wash in water of low salinity (Approx  0.5%NaCl) . 

Yoxon (1999) showed on Skye that the geology of the coastal zone influences the number of freshwater pools and the number of otters.  The Torridonian sandstone coastline has the highest density of otters with 1.1 otters per kilometre and the lowest number was found on the Tertiary Lavas and Intrusive with 0.12 otters per kilometre of coastline.  The Lewisian gneiss, the same rock as Barra, had a density of less than 0.16 otters per kilometre.  Using the estimate for Barra of 16-24 otters this would produce a figure between 0.3-0.5 otters per kilometre, which is higher than the Skye figure. 

REFERENCES 

Beja, P.R. (1991). Diet of otters (Lutra lutra) in closely associated freshwater-brackish and marine habitats in south west Portugal.   Journal of Zoological Society of London, 225, 141-152. 

Blatt, H., Middleton, G. and Murray, R.  (1972).  The Origin of Sedimentary Rocks.  Prentice Hall. 

Carss, D.N. (1996). Errors associated with otter (Lutra lutra,) faecal analysis. Assessing general diet from spraints. Journal of the Zoological Society, London, 238, 301-3 17. 

Conroy, J.W.H. and French, D.D. (1985). Monitoring otters in Shetland. Shell Oil Terminal Rport. 

Conroy, J.W.H. and French, D.D. (1987). The use of spraints to monitor populations of otters (Lutra lutra). Symposium of the Zoological Society of London, 58, 247-262. 

Conroy, J.W.H., Watt, I., Webb, J.B. and Jones, A. (1993). A guide to the identification of prey in otter spraint. Occasional publication 16, The Mammal Society, London. 

Erlinge, S.  (1967). Home range of the otter  (Lutra lutra) in southern Sweden. Oikos, 18.  379-97. 

Erlinge, S.  (1968). Territoriality of the otter (Lutra lutra). Oikos, 19, 8 1-98. 

Gore, P. (1997). Sedimentary Geology Lecture. World Wide Web. 

Jenkins, D., Walker, J.G.K. and McCowan, D. (1979).  Analysis of otter (Lutra lutra) faeces from Deeside, north east Scotland.  Journal of the Zoological Society of London, 187, 235-44. 

Jolly, G. M. (1969). The treatment of errors in aerial counts of wildlife populations. East African Agriculture and Forestry Journal, 34, 50-5 5. 

Heggberget, TM. (1993). Marine feeding otters (Lutra lutra) in Norway: Seasonal variation in prey and reproductive timing. Journal of the Marine Biological Association, 73, 297-312. 

Kruuk, H., Moorhouse, A., Conroy, J.W.H. and Durbin, L. (1989). An estimate of numbers and habitat preferences of otters (Lutra lutra) in Shetland Biological Conservation, 49, 241-254. 

Kruuk, H. and Balharry, D. (1990). Effects of sea water on thermal insulation of the otter (Lutra lutra). Journal of the Zoological Society of London, 220, 405-4 15. 

Kyne, M. J., Small, CM. and Fairley, J.S. (1989). The food of otters (Lutra lutra) in the Irish midlands and a comparison with that of mink (Mustela vison). Proceedings of the Royal Irish Academy, 89B, 33-46. 

Lovett, J.L. (1994). Utilisation of freshwater pools by the Eurasian otter in a marine environment. MSc Thesis, Aberdeen University. 

Lovett, J.L., Kruuk, H. and Lambin, X. (1997). Factors influencing the use of freshwater pools by otters (Lutra lutra) in a marine environment.  Journal of the Zoological Society of London, 75, 825-83 1 

Moorhouse, A. (1988). Distribution of holts and their utilisation by the European otter (Lutra lutra) in a marine environment. MSc Thesis, University of Aberdeen. 

Pond, C..M. and Mattacks, C.A. (1985). Anatomical organisation of mammalian adipose tissue.  Fortschritte der Zoologie, 30, 445-448. 

Priestley, S.H. (1996). The salinity of water pools used for bathing by the Eurasian otter (Lutra lutra). MSC thesis. DICE. 

Read, P. and Watson, J.  (1972).   Beginning  Geology.  .Macmillan and Co.

Rowe-Rowe, D.T. (1977).  Food ecology of otters in Natal, South Africa. Oikos, 28,2 10-219. 

Tarasoff, F.J. (1974). Anatomical adaptations in the River Otter, Sea Otter and Harp Seal with reference to thermal regulation. In Functional Anatomy of Marine Mammals, 2  (ed. R.J. Harrison), 111-142. Academic, London, 

van der Zee, D. (1981). Prey of the Cape Clawless Otter (Aonyx capensis) in the Tsitsikama Coastal National Park, South Africa. Journal of the Zoological Society of London, 194, 467-483. 

Watson, H. (1978). Coastal otters (Lutra lutra) in Shetland. Vincent Wildlife Trust. London. 

Wise, M.H., Linn, I.J. and Kennedy, C.R. (1981). Comparisons of the feeding biology of mink (Mustela vison) and otter (Lutra lutra).  Journal of the Zoological Society of London., 195, 181-213. 

Yoxon, P.  (1999).  The effect of geology on the distribution of the Eurasian otter (Lutra lutra) on the Isle of Skye.   Phd Thesis, Open University, Milton Keynes.