Counts of Puffins in Shetland Suggest an Apparent Decline in Numbers

Abstract

Scottish Birds: Journal of the Scottish Ornithologists' Club

Full text

223
Scottish Birds: 223–231
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)
Counts of Puffins in Shetland
suggest an apparent decline
in numbers
E. OWEN, O. PRINCE, C. CACHIA-ZAMMIT, R. CARTWRIGHT,
T. COLEDALE, S. ELLIOTT, S. HADDON, G.K. LONGMOOR,
J. SWALE, F. WEST & R. HUGHES
During May 2017, we counted Apparently Occupied Burrows (AOBs) of Puffins in a reference plot
at Hermaness National Nature Reserve, Shetland and found that numbers had declined by 69%
since 2002 (from 145 to 45 AOBs). Across the wider Hermaness site, counts of individuals were
adjusted using a correction factor based on the reference plot, and these indicated that declines of
42% have occurred. Counts made of individuals on land in 29 areas of Shetland (including
Hermaness) in May, where burrows were not accessible, were lower than anticipated but valid
comparison with past counts is not possible since these had been made in June when increasing
numbers of non-breeders attend colonies. Taken together with findings from other studies in
Shetland, our results suggest that Shetland’s Puffin population has declined.
Introduction
Puffin Fratercula arctica breeding failures have been seen in Iceland and Norway (where 80%
of the world’s Puffins breed; BirdLife International 2015), and further declines in numbers are
anticipated, leading to the species being declared Vulnerable to extinction on the IUCN Red List
(BirdLife International, 2015). These breeding failures are often attributed to global warming
causing oceanographic changes that result in changes in Puffin prey distribution and
abundance (Durant et al. 2003). Around 10% of the world’s Puffins breed around the coast of
Britain and Ireland, with the majority in Scotland (82%), followed by England (13%) and
smaller populations in Ireland (4%) and Wales (2%) (Mitchell et al. 2004). Shetland has histor-
ically held 14–17% of the Puffins breeding in Britain and Ireland (Lloyd et al. 1991, Mitchell
et al. 2004). The Shetland Puffin population has been assessed three times. Operation Seafarer
(1969–70) recorded 65,054 Apparently Occupied Burrows (AOBs), the Seabird Colony Register
(1985–88) 104,381 AOBs, and Seabird 2000 (1998–2002) 107,676 AOBs. These suggested an
upward population trend of 66% between 1970 and the mid-1980s, slowing to just 3% between
the mid-1980s to the turn of the century (Mitchell et al. 2004). Since then considerable
decreases have been observed in other seabirds (Mavor et al. 2008) and in the three Puffin
colonies in Shetland that have been counted more recently (Table 1; Miles et al. 2015).
Table 1. Recent Puffin population estimates at three colonies in Shetland compared to estimates in 1999–2000. All
estimates are based on counts of individuals and were carried out in May (Noss); late April (Fair Isle, and Foula in 2016);
or early June (Foula, 2000). *Counts suggest declines on all three sites though caution should be used when interpreting
population change using counts of individuals, particularly on Foula where counts were made in different months.
Colony Most recent count Count in 1999–2000 % change References
(year in brackets) (year in brackets) between
counts*
Noss 1,174 (2017) 1,892 (1999) -37.9 Nisbet and Denton, 2017; Mitchell et al. 2004
Fair Isle 6,666 (2015) 15,000 (2000) -62.5 Parnaby et al. 2017, Miles et al. 2015
Foula 5,055 (2016) 22,500 (2000) -77.5 SNH/Foula Ranger Service Unpublished data,
Mitchell et al. 2004

Puffins are difficult to census because of their burrow nesting habits and often inaccessible
breeding locations. The most reliable method is by long-term monitoring of AOBs in sample areas
of a colony (Walsh et al. 1995) but this requires access to burrows, which is rarely possible in
Shetland due to the steep and inaccessible terrain (Mitchell et al. 2004). When burrows cannot be
accessed, the standard protocol is to count birds that are on land (Walsh et al. 1995). These counts
provide an estimate, which could differ from the true population size by an order of magnitude
(Walsh et al. 1995, Calvert & Robertson 2002) but such counts are useful to indicate change in
populations where those changes are large and over longer time periods (Miles et al. 2015).
Standard monitoring methods (Walsh et al. 1995, Mitchell et al. 2004) advise that these counts of
individual birds are best obtained during the pre-laying period (mid-April to early May in
Shetland) but if counts cannot be made at this stage then they should be obtained before non-
breeding birds begin attending the colony, which in Shetland is usually in June (Harris & Wanless
2011). However, because the Shetland coastline is long and inaccessible, Puffins in the majority
of locations have normally been counted during June at the same time as the optimal count
period for other cliff-nesting seabirds (though Fair Isle and Noss have been counted during
April/May). These previous surveys in June will result in counts that are inflated by the presence
of non-breeding birds. We counted Puffins at key colonies on mainland Shetland, Yell and Unst,
in order to estimate the breeding populations at these sites. We compare our counts with Seabird
2000 or other previous surveys and discuss possible changes in Puffin populations in Shetland.
Materials and Methods
The largest colony in our sample was at Hermaness National Nature Reserve, hereafter ‘Hermaness’.
This colony has been previously counted using a site-specific method in June (Martin 1995, 1997,
2002) which attempts to account for the fact that so few of the burrows at this site are visible due
to Puffins nesting in boulders, a method which has also been used on The Shiants (Brooke 1972).
In this method, a reference plot at Sothers Brek (Plate 185), where AOBs had previously been
counted, was used to calculate a ratio of visible birds on land: AOBs every five minutes, while the
wider site is counted by surveyors. It was difficult to relocate this reference plot (Plate 185, blue line)
at Sothers Brek, which meant the area we surveyed for the reference plot was later realised to be
slightly larger than the original reference plot (Plate 185, gold line). We counted the number of AOBs
in the reference plot on 22 May 2017. On 28 June, 11 people simultaneously surveyed allotted
sections of the whole Hermaness site (c. 8 km in total length), over the course of an hour recording
the time of count sections into five-minute bands while a twelfth counter counted visible birds at
the reference plot every five minutes. Martin (1995, 1997, 2002) did this to attempt to account for
changes in loafing numbers over the course of the survey. The ratio for the corresponding five-
minute interval was then applied to the timed counts of birds on land across the whole Hermaness
colony. Following Martin (1995, 1997, 2002) we multiplied the number of visible birds on land
around the reserve in every five-minute band by the number of AOBs previously counted in the
reference plot and divided by the number of birds visible in the plot at the same time of evening
under the assumption that loafing rates are consistent across the wider Hermaness site. For one of
our five-minute intervals the count of birds on land within the reference plot was zero, precluding
the calculation of a meaningful ratio. Therefore, for this time interval we averaged the number of
visible birds in the reference plot across the previous and next five-minute interval to obtain a non-
zero ratio. To be consistent with Martin (1995, 1997, 2002) we added 10% to the adjusted count to
correct for birds on land assumed to have been missed (due to the cliff terrain).
Separately, we counted Puffins during May at selected colonies on Shetland, including
Hermaness, prioritizing larger colonies, as measured in Seabird 2000. We used the same stretches
of coastline (‘count sectors’) as had been used in Seabird 2000 and all data were submitted to the
JNCC’s Seabird Monitoring Programme. No count sectors were accessible along their entirety for
burrow inspection and therefore counts of individuals on land were used. In addition, separate
counts were made of Puffins sitting on the sea within 200 m of land and flying.
Scottish Birds: 223–231
224
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)

225
Scottish Birds: 223–231
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)
Counts were conducted from land at vantage points along the coastline of each count sector,
taking care to ensure that the whole census sector had been counted while minimising double-
counting of individuals (both within and between sectors). All counters had been trained in Puffin
census methods by an experienced Puffin surveyor (RH) and used binoculars (8x or 10x magnifi-
cation) to count Puffins within a clearly viewable range (varied depending on conditions but not
more than 300m). Seabird 2000 prescribed that Puffins be counted during ‘daylight’ (Mitchell et
al. 2004) while Walsh et al. (1995) recommend that counts are done in the evening when the
number of individuals around the colony are usually highest. We aimed to do counts in the
evening, but because of logistical (e.g. available ferries) and weather constraints (poor visibility
(<200 m due to fog); persistent rain; or wind above Beaufort scale 4) we also made counts during
the day. Where time allowed we counted sectors more than once, at least seven hours apart.
Plate 185. Sothers Brek reference plot, Hermaness, Shetland, in 1988 (blue) and 2017 (gold). © Robert Hughes
Results
Hermaness
Weather conditions were calm at Hermaness on 28 June 2017 and attendance at the colony
seemed high relative to attendance seen since 2015 (Rachel Cartwright pers. obs.), which suggests
our count was conducted when colony attendance was relatively high and the count was made
prior to dusk, as had been done in previous years. A total of 45 AOBs were counted in the
reference plot. The number of birds seen on land at the reference plot was generally lower than
the number of AOBs present. We counted 3,588 birds on land over the whole count sector which,
when adjusted by the ratio from the reference plot, equates to an estimate of 12,521 AOBs (Table
2). A total of 3,043 birds on the sea and 1,069 flying birds were counted. For Hermaness, it was
possible to calculate percentage change in numbers since previous surveys because the same
methodologies and months of survey were used for each (Table 3). The number of AOBs in the
reference plot decreased by 69% since 2002 and 79% since 1997. Across the Hermaness count
sector, the unadjusted counts of birds on land decreased by 77% between 1997 and 2017 and the
adjusted counts decreased by 42% since 2002 and 58% since 1997.

Scottish Birds: 223–231
226
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)
Figure 1. Locations of Puffin count sector start points in Shetland, 2017.
1. Ayre of Tonga to Tonga Stack
2. Birrier to Rivalee
3. Broad Stack to Stack of Barons Geo
4. Burn of Garth to Broad Stack
5. Burrafirth to The Keen
6. Fugla Geo
7. Geo of Vigon to Birrier
8. Greff to Ayre of Tonga
9. Hermaness NNR
10. Hols Hellier
11. Kame to Corbie Geo
12. Landvillas to Scarfi Taing
13. Norwick Herda
14. Rivvalee to Wester Lee of Gloup
15. Scarfi Taing to Noup o’ Noss
16. Skitstack to Woodwick
17. Snarra Voe to The Keen
18. South Geo Brough to Greff
19. Stack o’ da Noup to Kame
20. Stack of Barons Geo to The Nev
21. Sumburgh Head
22. The Kame to Landvillas
23. The Keen to Lunda Wick
24. The Lug
25. The Nev to The Kame
26. Uyea
27. Virdick
28. Wick of Collaster to Skitstack
29. Woodwick to South Geo Brough

227
Scottish Birds: 223–231
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)
Elsewhere
In total, 50 counts of 29 count sectors were conducted on 18 days between 2 May and 24 May
2017 (Table 4; Figure 1), with 16 count sectors having more than one count.
Counts of birds on land were generally low with 14 out of 29 completed count sectors having
no Puffins on land (Table 4) and two sectors had no Puffins on land, sea or flying. As expected,
repeat counts often varied, with the most marked variation being at Sumburgh Head where the
number of birds on land ranged from 20 to 474 individuals over three counts. Of the 4,728
Puffins counted during May (including multiple counts at the same colony) the majority (71%;
3,371 individuals) were counted on the water with far fewer counted on land (16%; 735
individuals) or flying (13%; 622 individuals).
It is not possible to make a valid comparison with past counts since these had been made in
June when increasing numbers of immatures attend the colony, so counts made then are likely
to be substantially higher than those made in May (Harris & Wanless 2011). For instance, we
counted only 20 birds ashore at Hermaness on both 21 and 23 May but 3,588 on 28 June
(Tables 1 and 4). However, none of the 29 completed count sectors in 2017 had more Puffins
than were reported in Seabird 2000 (Table 5). We counted 578 Puffins on land during our May
counts, compared to 32,133 Puffins on land counted in these sectors during June in Seabird
2000. If we exclude the largest site, Hermaness, then our May counts for Puffins on land total
558 Puffins, compared to 8,472 during Seabird 2000.
Table 2. Hermaness-specific colony size estimation following Martin (1995, 1997, 2002). Colony-wide counts on 28
June of birds on land, sea and flying and birds visible on land within a reference plot previously determined to have
45 AOBs, recorded in five-minute bands.
Time (BST) Total
22:30 22:35 22:40 22:45 22:50 22:55 23:00 23:05 23:10 23:15 23:20 23:25
Land 178 900 644 145 33 116 56 317 39 662 140 358 3,588
Sea 590 166 446 0 1,350 34 39 401 0 17 0 0 3,043
Flying 188 316 142 84 52 42 60 55 10 68 7 45 1,069
Birds on land at
reference plot 21 19 10 0 14 39 43 4 38 13 45 72
AOBs in reference plot 45 45 45 45 45 45 45 45 45 45 45 45
Adjusted land count 381 2,132 2,898 544* 106 134 59 3,566 46 2,292 140 224 12,521
*birds on land at reference plot - value of 12 used as an average of previous and following counts (see methods)
Table 3. Puffin numbers at Hermaness (1995–2017) in Sothers Brek reference plot (AOBs) and over whole reserve
(unadjusted birds on land and adjusted birds on land), together with percentage changes between successive counts.
Data taken from Martin (1995, 1997 and 2002) and this study (2017).
Birds on land - Birds on land -
Year AOBs in plot whole reserve unadjusted whole reserve adjusted
Number Percentage Unadjusted Percentage Adjusted Percentage
change change change
1995 147 unavailable 22,000
1997 218 +48.3 15,600 28,300 +29
2002 145 -33.5 unavailable 23,661 -16
2017 45 -69.0 3,588 -77.0 13,773 -42

Scottish Birds: 223–231
228
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)
Table 4. Number of Puffins counted at 29 count sectors in Shetland during May 2017 (see Figure 1).
Site number Site Name Date Time Land Sea Flying
1 Ayre of Tonga to Tonga Stack 18 May 17 13:10–13:43 0 2 6
Ayre of Tonga to Tonga Stack 19 May 17 15:30–15:39 0 59 3
2 Birrier to Rivalee 11 May 17 16:05–16:34 0 10 0
3 Broad Stack to Stack of Barons Geo 4 May 17 10:55–11:12 0 24 1
Broad Stack to Stack of Barons Geo 6 May 17 11:22–11:35 0 0 0
Broad Stack to Stack of Barons Geo 8 May 17 12:25–12:43 0 0 0
4 Burn of Garth to Broad Stack 4 May 17 11:15–12:03 2 4 1
Burn of Garth to Broad Stack 6 May 17 09:43–11:16 0 0 0
Burn of Garth to Broad Stack 8 May 17 09:43–12:16 0 1 1
5 Burrafirth to The Keen 22 May 17 09:50–10:52 0 0 0
6 Fugla Geo 15 May 17 06:49–08:42 0 2 0
Fugla Geo 15 May 17 17:12–19:17 0 0 0
7 Geo of Vigon to Birrier 11 May 17 16:36–16:58 0 0 0
8 Greff to Ayre of Tonga 18 May 17 11:58–12:27 0 79 3
Greff to Ayre of Tonga 19 May 17 15:30–15:39 0 22 2
9 Hermaness NNR 21 May 17 19:00–20:22 20 261 36
Hermaness NNR 23 May 17 20:30–22:00 20 1,569 78
10 Hols Hellier 24 May 17 10:05–10:36 2 92 8
11 Kame to Corbie Geo 5 May 17 10:32–10:45 5 2 1
12 Landvillas to Scarfi Taing 4 May 17 14:36–15:10 0 11 2
Landvillas to Scarfi Taing 5 May 17 10:05–10:45 7 79 27
13 Norwick Herda 24 May 17 10:06–11:18 0 7 0
14 Rivvalee to Wester Lee of Gloup 11 May 17 15:08–16:03 0 24 0
Rivvalee to Wester Lee of Gloup 15 May 17 17:13–18:14 1 8 0
15 Scarfi Taing to Noup o’ Noss 4 May 17 13:43–14:36 36 130 11
16 Skitstack to Woodwick 18 May 17 09:24–11:09 1 8 1
17 Snarra Voe to The Keen 22 May 17 11:28–12:27 0 3 0
Snarra Voe to The Keen 19 May 17 11:22–12:20 2 6 10
18 South Geo Brough to Greff 18 May 17 10:20–11:58 0 48 10
19 Stack o’ da Noup to Kame 4 May 17 13:28–13:38 0 1 0
Stack o’ da Noup to Kame 5 May 17 10:13–10:32 8 71 1
20 Stack of Barons Geo to The Nev 2 May 17 18:34–19:44 2 380 16
Stack of Barons Geo to The Nev 10 May 17 11:05–12:27 0 4 3
21 Sumburgh Head 2 May 17 18:13–19:28 20 13 37
Sumburgh Head 3 May 17 09:35–11:05 117 73 88
Sumburgh Head 4 May 17 09:00–10:45 474 84 242
22 The Kame to Landvillas 4 May 17 15:10–16:44 0 18 2
The Kame to Landvillas 5 May 17 10:45–11:04 1 38 1
The Kame to Landvillas 8 May 17 12:30–12:50 0 1 0
23 The Keen to Lunda Wick 17 May 17 16:40–17:31 0 0 0
The Keen to Lunda Wick 19 May 17 10:04–11:20 0 0 8
The Keen to Lunda Wick 22 May 17 10:31–11:27 0 0 1
24 The Lug 24 May 17 10:36–11:14 0 100 8
25 The Nev to The Kame 2 May 17 18:40–19:45 9 78 1
The Nev to The Kame 8 May 17 11:40–12:30 0 0 0
26 Uyea 9 May 17 14:19–15:09 8 33 10
27 Virdick 24 May 17 10:18–11:09 0 8 0
28 Wick of Collaster to Skitstack 18 May 17 11:10–12:10 0 11 0
29 Woodwick to South Geo Brough 17 May 17 17:00–18:18 0 7 0
Woodwick to South Geo Brough 18 May 17 09:22–10:20 0 0 3

229
Scottish Birds: 223–231
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)
Discussion
Our survey was not a complete census of the total Shetland Puffin population. However, the count
sectors that we completed held 30% of the Puffin population of Shetland at the time of the last
census. With the addition of the large colonies listed in Table 1 (Noss, Fair Isle and Foula), which
were recently counted by Scottish Natural Heritage, Fair Isle Bird Observatory and Foula Ranger
Service, respectively, we calculate recent counts have covered an area which held 90% of
Shetland’s Puffins during Seabird 2000.
Counting Puffins is challenging and this is especially true in the rugged terrain of Shetland. We used
three methods to assess population change, all with different levels of certainty associated with them
(Table 6). The most accurate way of assessing change in Puffin population size is through counts of
AOBs. Our repeat count of AOBs at one such plot at Hermaness indicated a decline of 69% since 2002.
All other counts were made of individual birds on land at a colony. Such counts can misrepresent the
true population size by up to an order of magnitude because attendance changes with season, time
of day, weather and due to the presence of predators (Walsh et al. 1995, Calvert & Robertson 2002).
Therefore, these counts are only useful to indicate large population changes.
Table 5. Puffin counts at 29 count sectors in Shetland during Seabird 2000 and 2017. Where multiple counts were
made in 2017, the highest count is shown. Seabird 2000 data downloaded from www.jncc.defra.gov.uk on 10 June
2017. * Hermaness specific method used
Seabird 2000 (1999–2002) June May 2017 - unlikely
- likely to include non-breeders to include non-breeders
Subsite Land Sea Land Sea
Ayre of Tonga to Tonga Stack 70 700 0 59
Birrier to Rivalee 28 121 0 10
Broad Stack to Stack of Barons Geo 40 unavailable 0 24
Burn of Garth to Broad Stack 536 unavailable 2 4
Burrakirk to the Keen unavailable 24 0 0
Fugla Geo 2 unavailable 0 2
Geo of Vigon to Birrier 13 75 0 0
Greff to Ayre of Tonga 77 unavailable 0 79
Hermaness NNR *23,661 unavailable 20 1,569
Hols Hellier 280 unavailable 2 92
Kame to Corbie Geo 92 unavailable 5 2
Landvillas to Scarfi Taing 846 unavailable 7 79
Norwick Herda 950 95 0 7
Rivvalee to Wester Lee of Gloup 193 unavailable 1 8
Scarfi Taing to Noup o’ Noss 506 1600 36 130
Skitstack to Woodwick 66 44 1 8
Snarra Voe to The Keen 31 18 2 6
South Geo Brough to Greff 414 169 0 48
Stack o’ da Noup to Kame 560 unavailable 8 71
Stack of Barons Geo to The Nev 117 unavailable 2 380
Sumburgh Head 502 unavailable 474 84
The Kame to Landvillas 788 unavailable 1 38
The Keen to Lunda Wick 4 unavailable 0 0
The Lug 700 unavailable 0 100
The Nev to The Kame 8 unavailable 9 78
Uyea 925 250 8 33
Virdick 625 unavailable 0 8
Wick of Collaster to Skitstack 53 20 0 11
Woodwick to South Geo Brough unavailable 22 0 7
Total land count (if no land count, sea is used) 32,133 578

The site-specific method we repeated at the wider Hermaness site in June relies on such counts of
individuals and these raw counts suggest a decline of 77%. Once the adjustment factor is applied,
a smaller decline of 51% since 1997 or 42% since 2002 is indicated. However, the reliability of
adjusting counts to provide a population estimate using an index of visible birds to AOBs across
the wider site is untested. The use of separate adjustment factors for each five-minute interval
assumes that the ratio of birds on land: AOB is reliable for the entire 8 km coastline on a five-
minute basis. The size of our adjusted counts is more dependent on the magnitude of the five-
minute adjustment factors than on the size of the raw counts themselves. This in itself is not of
concern, if the adjustment factors are reliable but as they are highly variable (ranging from 0.625
to >11) and used to adjust counts up to 8km away, this may not be the case. In previous surveys
the correction factors may have been less variable between the five-minute intervals since the
calibration plot provided a larger sample size (150–210 AOBs). In the future, a larger calibration
plot (more AOBs) may be helpful in reducing the variability in the adjustment factors.
All remaining sectors were counted using counts of individuals on land in May for which there
is no directly comparable count in previous censuses (as they were done in June), though the
broad scale differences to the Seabird 2000 counts were marked (Table 4). Some of the difference
between Seabird 2000 and 2017 will be due to methodological differences, in particular whether
non-breeders were included in the count. Harris (1984) estimated between 10 and 20% of the
population comprises immatures during chick-rearing, based on age classes determined by bill
grooves. Our 2017 counts at Hermaness suggested that the total number of birds on land and on
the sea was four times larger in June (6,631) than in May (1,589) and the birds on land figure on
its own was 179 times greater in June (3,588) than in May (20). On Noss, May and July counts
were carried out in 1994, 1999 and 2000 and the July counts were respectively 1.8, 2.2 and 4.9
times greater than May counts (SNH unpublished data). Despite this uncertainty, the body of
evidence presented here along with the 38%, 78% and 63% declines on, respectively, Noss, Foula
(but note that counts here were made in different months) and Fair Isle (Table 1) support the
suggestion that large declines are likely to be widespread across Shetland’s Puffin colonies.
Acknowledgements
We are grateful to Afra Skene, Sheila Gear and David Parnaby for providing information on Puffin
counts on Noss, Foula and Fair Isle and Mike Harris, Sarah Wanless, and Tony Martin for
discussion of Puffin census methodology. We thank landowners for access to Puffin colonies.
SNH, RSPB and SOTEAG staff on Shetland for their help in co-ordinating which sites to count
and for joint discussion of an earlier draft of this report, which greatly improved it. In particular,
Martin Heubeck and Helen Moncrieff provided additional information for some sites. Mark Bolton
and Euan Dunn and two anonymous referees provided comments to improve the paper. Counts
were carried out as part of Project Puffin (UK), funded by RSPB and the Heritage Lottery Fund.
Scottish Birds: 223–231
230
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)
Table 6. Direction and strength of evidence of population changes associated with three census methods used in
Puffin counts in Shetland during 2017.
Census method Result Strength of evidence
Reference Plot -69% since 2002 Strong evidence of decline since 1997
count (Hermaness) -79% since 1997 but in small area (<200 AOBs)
Hermaness-specific Adjusted estimates: Medium evidence of decline using
(June) -42% since 2002 method consistent, but non-standard,
-51% since 1997 method since 1997
Raw counts of birds on land: -77% since 1997
Individuals on land at Low overall numbers and 14 of 29 count Suggestion of widespread decline, but
29 count sectors (May) sectors with no Puffins on land. survey months inconsistent between years.

231
Scottish Birds: 223–231
Counts of Puffins in Shetland suggest an apparent decline in numbers
38:3 (2018)
References
BirdLife International. 2015. European Red List of Birds. Office for Official Publications of the
European Communities, Luxembourg.
Brooke, M. de L. 1972. The Puffin population of the Shiants. Bird Study 19: 1–6.
Calvert, A.M. & Robertson, G.J. 2002. Colony attendance and individual turnover of Atlantic
Puffins in Newfoundland. Waterbirds 25(3): 382–387.
Durant, J.M., Anker-Nilssen, T. & Stenseth, N.C. 2003. Trophic interactions under climate
fluctuations: the Atlantic puffin as an example. Proceedings of the Royal Society of London.
Series B Biological Sciences 270(1523): 1461–1466.
Harris, M.P. 1984. The Puffin. T. & A.D. Poyser, Staffordshire.
Harris, M.P. & Wanless, S. 2011. The Puffin. Bloomsbury Publishing.
Martin, A.R., 1995. Ornithological Fieldwork on Hermaness NNR, Shetland conducted by the
Cambridge Group, June 1995. Unpublished report to NCC and SOTEAG.
Martin, A.R., 1997. Ornithological Fieldwork on Hermaness NNR, Shetland conducted by the
Cambridge Group, June 1997. Unpublished report to NCC and SOTEAG.
Martin, A.R., 2002. Ornithological Fieldwork on Hermaness NNR, Shetland conducted by the
Cambridge Group, June/July 2002. Unpublished report to JNCC and SOTEAG.
Mavor, R.A., Heubeck, M., Schmitt, S. & Parsons, M. 2008. Seabird numbers and breeding
success, 2006. UK Nature Conservation No. 31.
Miles, W.T.S., Mavor, R., Riddiford, N.J., Harvey, P.V., Riddington, R., Shaw, D.N., Parnaby, D.
& Reid, J.M. 2015. Decline in an Atlantic Puffin Population: Evaluation of Magnitude and
Mechanisms. PLoS ONE 10(7): e0131527.
Mitchell, P.I., Newton, S.F., Ratcliffe, N. & Dunn, T.E. 2004. Seabird populations of Britain and
Ireland. T. & A.D. Poyser, London.
Nisbet, C. & Denton, A. 2017. Noss NNR Annual Report. Scottish Natural Heritage, Lerwick.
Parnaby, D., Hatsell, C.R., Cope, R. & Dodd, C.J. 2017. Fair Isle Seabird Studies 2017. Joint
Nature Conservation Committee, Aberdeen.
Walsh, P.M., Halley, D.J., Harris, M.P., del Nevo, A., Sim, I.M.W. & Tasker, M.L. 1995. Seabird
monitoring handbook for Britain and Ireland. JNCC/RSPB/ITE/Seabird Group, Peterborough.
Ellie Owen, Oliver Prince, Chris Cachia-Zammit, Sophie Elliott, Sian Haddon, Georgia K
Longmoor, Fritha West & Robert Hughes, RSPB Centre for Conservation Science, RSPB,
Etive House, Beechwood Park, Inverness IV2 3BW.
Rachel Cartwright & Jonathan Swale, SNH, Great Glen House, Leachkin Road,
Inverness IV3 8NW.
Tessa Coledale, RSPB Centre for Conservation Science, RSPB, 2 Lochside View,
Edinburgh Park, Edinburgh EH12 9DH.
Revised ms accepted June 2018