NatureScot Research Report 1331 - An analysis of Scottish Raptor Monitoring Scheme data to improve understanding of the impact of the 2022 Avian Influenza outbreak on raptors and raptor populations in Scotland

Year of publication: 2023

Authors: Wilson, M.W., Wetherhill, A.S., and Wernham, C.V. (British Trust for Ornithology)

Cite as: Wilson, M.W., Wetherhill, A.S., and Wernham, C.V. 2023. An analysis of Scottish Raptor Monitoring Scheme data to improve understanding of the impact of the 2022 Avian Influenza outbreak on raptors and raptor populations in Scotland. NatureScot Research Report 1331.

Keywords

raptor, Scotland, highly pathogenic avian influenza, HPAI, bird flu, breeding success

Background

The outbreak of Highly Pathogenic Avian Influenza (HPAI) that began in 2021 has impacted a wide variety of species in the UK (Pearce-Higgins et al., 2023). Impacts in breeding seabird colonies and wintering goose populations were the most obvious examples, but testing showed that, by the spring of 2022, HPAI was infecting a wide variety of wild bird species in Scotland, including raptors (Animal and Plant Health Agency, 2023). Because raptors are more dispersed and often inhabit more remote locations, there was concern that the impact of HPAI in these birds was being underestimated. In order to address this issue, data from the Scottish Raptor Monitoring Scheme (SRMS) collected during the breeding season of 2022 was used to compare the breeding performance of common buzzard, golden eagle, goshawk, hen harrier, kestrel, merlin, osprey, peregrine falcon, raven, red kite, sparrowhawk and white-tailed eagle with comparable data from previous years. The objective was to assess whether significant changes in numbers of breeding birds or in breeding success had occurred and whether any of the changes detected were likely to be caused by the HPAI outbreak.

Main findings

There is strong evidence for declines in breeding success consistent with impacts of avian influenza on the breeding productivity of golden eagle and white-tailed eagle in 2022.
Across the whole SRMS dataset, golden eagle breeding success (the percentage of territorial pairs that successfully reared young) declined from 48% to 28%. White-tailed eagle breeding success declined from 67% to 45%. The largest declines recorded for both species were in Lewis & Harris, where breeding success of golden eagles fell from 55% to 16%, and breeding success of white-tailed eagle declined from 66% to 24%.
These impacts are evident in most of the regions where these species breed but, for both species, they appear to be greater in areas where breeding pairs have access to coastal and marine habitats, indicating a possible link to predation and scavenging of infected seabirds and waterfowl. White-tailed eagle breeding success in three coastal study areas experienced statistically significant declines from between 60% to 67% before 2022 to between 16% and 50% in 2022. By contrast, breeding success of inland breeding pairs during the same period rose (non-significantly) from 67% to 77%.
Impacts on the breeding success of other raptor species do not appear to have been so widespread or severe, but the evidence is suggestive of some other local impacts on the proportion of pairs successfully fledging offspring. This evidence includes statistically significant declines in breeding success of ravens in Shetland and Orkney, peregrines in Tayside and red kites in Highland North-east Scotland.
Impacts on the number of pairs attempting to breed in 2022 are not so evident from SRMS monitoring outcomes. Among study populations for which the number of home ranges checked in 2022 was similar to that in previous years, fewer pairs were recorded in 2022 than in previous years for buzzards in Tain, Ross-shire and for hen harriers on Rousay, Orkney.
The number of offspring fledged by successfully breeding pairs in 2022 appears to be broadly similar to that in previous years for most species in most areas.
The fact that breeding success appears to have been more severely affected than fledged brood size suggests that where HPAI results in nestling mortality, it is relatively uncommon for any chicks from affected broods to survive.
Other factors that could explain the differences observed between 2022 and other years, particularly in breeding success, include variation in weather, prey availability and survey effort. Of these, the weather recorded in 2022 may have contributed to the observed differences, but seems unlikely to entirely account for all of them.
Anecdotal observations by observers include many instances where breeding failure was suspected to have been caused by HPAI. In general, these observations fit well with the patterns described above, including breeding failures of golden eagle, white-tailed eagle, buzzard, red kite and raven pairs.

Acknowledgements

We are grateful to the Scottish Raptor Monitoring Scheme (SRMS) for providing the data on which this report is based, and in particular to its coordinator, Amy Challis, and the partner organisations who support the SRMS and approved access to these data. Huge thanks go to the hundreds of dedicated raptor surveyors who contributed to the SRMS dataset, especially to those who responded directly to our request for information on possible HPAI impacts from their own observations in 2022. The study was funded by NatureScot, and benefited from helpful comments and suggestions from John Allan, Andrew Stevenson and David Jardine.

Abbreviations

HPAI – Highly Pathogenic Avian Influenza

SRMS – Scottish Raptor Monitoring Scheme

BTO – British Trust for Ornithology

RSPB – Royal Society for the Protection of Birds

JNCC – Joint Nature Conservancy Council

BBS – Breeding Bird Survey

NRS – Nest Record Scheme

SGAR – Second Generation Anticoagulant Rodenticide

Terminology

The following terminology is used in this report:

breeding success refers to the proportion of pairs which successfully fledge one or more offspring.

fledged brood size refers to the number of young successfully reared by pairs that fledged at least one young.

breeding productivity is used more generically to refer to the combined effects of breeding success and fledged brood size (i.e. the overall number of offspring reared by breeding pairs).

1. Introduction and aims

An outbreak of Highly Pathogenic Avian Influenza (HPAI) in wild birds in Scotland, as well as the wider UK, has negatively impacted several species of wild birds (Pearce-Higgins et al., 2023). While seabirds and wildfowl have been the most visibly affected groups, other groups have also been impacted, though the population consequences are not yet well understood. The first raptor cases detected in Scotland were in mid-November 2021 (Animal and Plant Health Agency, 2023). More detections of infection in raptors over the winter revealed that AI was circulating widely across much of the mainland. By mid-April 2022, positive tests for HPAI had been reported for white-tailed eagle, buzzard, red kite, peregrine, kestrel, sparrowhawk and golden eagle (Animal and Plant Health Agency, 2023). Based on these data, it seems likely that HPAI could have increased adult mortality and/or reduced breeding success of several raptor species during 2022.

Scottish Raptor Monitoring Scheme (SRMS) data from the breeding season of 2022 was compared with data from previous years in an effort to better understand the impact of HPAI on the numbers and productivity of Scotland’s Raptors. The objective was to highlight any anomalies in breeding numbers or performance that suggested impacts of HPAI. Comparing data for individual species drawn from discrete study areas and regions allowed an assessment to be made of the extent and severity of such impacts in different parts of Scotland.

This information can inform the decision-making of NatureScot and other organisations about monitoring and protection of raptor populations of conservation importance, evaluating the need for measures to safeguard both humans and raptors in the context of future raptor monitoring and other types of fieldwork, and developing recovery plans for these populations in the future should this prove necessary.

The work had two main objectives:

To quantitatively compare levels of breeding success recorded for raptors in 2022 with those levels recorded in previous years.
To summarise and qualitatively assess information reported by contributors to the Scottish Raptor Monitoring Scheme that may describe HPAI occurrence in, or impacts on, monitored raptor populations.

2. Methods

2.1 Scottish Raptor Monitoring Scheme data

The Scottish Raptor Monitoring Scheme (SRMS) covers 19 species of raptor known to breed regularly in Scotland, plus the raven (included as an ‘honorary raptor’ on the grounds of its ecologically similarity to some raptor species). Breeding raptor records submitted to the SRMS comprise information about the annual occupancy status of ‘home ranges’ (areas known to have been occupied by breeding raptors), and the breeding outcomes of any nesting attempts found in these home ranges. Occupancy status describes whether or not home ranges were checked and, if they were, whether there was any evidence of occupation by any territorial birds. For home ranges found to be occupied by territorial pairs, a number of metrics are recorded, where feasible, that describe various breeding stages and outcomes. These are: the number of eggs laid, the number of eggs hatched, the number of large chicks (well-feathered and/or sufficiently well-grown to be ringed), breeding success and the number of young known to have fledged. Records from each year are standardised and cleaned to make sure that the information they contain is both reliable and, as far as possible, comparable across regions and years. SRMS Annual Reports (e.g. Challis et al., 2022) summarise the information relating to the number of breeding attempts recorded and their breeding productivity by species for each of 12 SRMS regions. The metrics summarised in these reports include the number of breeding pairs recorded, breeding success (the proportion of these pairs fledging one or more offspring) and fledged brood size (the average number of young recorded as being reared from successful breeding attempts). Since 2018, these summaries have been restricted to records for which the contributors have signed up to SRMS’s Data-Sharing Policy, giving explicit permission for their information to be used and shared in this way. At the time of writing, SRMS had published the annual reports for all years up to 2020. For the current project, data from 2021 and 2022 were prepared in a manner comparable to previous years (using only records covered by the Data-Sharing Agreement) and used to produce equivalent summary tables of occupancy and breeding productivity with which to compare with information from previous years

In this report, breeding numbers and performance are evaluated for buzzard, golden eagle, goshawk, hen harrier, kestrel, merlin, osprey, peregrine, raven, red kite, sparrowhawk and white-tailed eagle. These comprise 12 of the 14 species for which recent SRMS data were considered sufficient to analyse inter-annual changes in these metrics (Wilson et al., 2022). Annual changes in breeding numbers and performance of the other two species, barn owl and tawny owl, were considered too volatile for them to be suitable for detection of impacts of HPAI within a single year. This is because the breeding performance of these species is highly dependent on the abundance of small mammal prey, which undergoes dramatic changes from year to year.

2.2 Local Studies

During preparation for the recently published SRMS report on raptor population trends (Wilson et al., 2022), local study areas were defined to ensure that comparisons between years were not skewed by temporal variation in survey effort. This is particularly important for comparisons of numbers of breeding pairs, which would otherwise be directly influenced by survey effort. Preliminary study areas were first identified from clusters of home ranges for which records had been submitted to the SRMS in multiple years. These clusters indicated areas within which survey effort was relatively high and consistent. Study areas were then refined on the basis of feedback from regional Scottish Raptor Study Group Species Coordinators, excluding areas with inconsistent coverage, or adding back in additional areas where search effort was known to be high even when the number of records reported was low (Wilson et al., 2022).

For the current analyses, the number and location of home ranges checked in 2022 within each defined study area were compared visually (in ArcGIS Desktop 10.6.1) with the number and distribution of records from the previous four years. If the number of ranges checked in 2022 (and the average number of ranges checked during the previous four years) was 10 or more, and the distribution of these home ranges suggested that monitoring coverage of the study area had remained broadly stable over time, the study area was used in our analyses. Where coverage in 2022 (as indicated by the distribution of 10 or more checked home ranges) appeared stable within part of a study area, but decreased elsewhere, study areas were ‘trimmed’ to exclude areas with low monitoring effort in 2022 before being included in our analyses. An additional area covering inland white-tailed eagle territories on the Scottish mainland that were not included in other study areas was also drawn in ArcGIS, in order to facilitate a comparison between inland and coastal home ranges.

For each study area, generalised linear models (GLMs) were used to evaluate whether the measures of breeding population size or productivity were lower in 2022 than in earlier years, as this would indicate a possible impact of the HPAI outbreak. Poisson GLMs were used to evaluate variation in the number of breeding pairs and fledged brood sizes, while binomial GLMs were used to analyse the proportion of pairs recorded as successfully producing young. For each of these three metrics, a single GLM was run for each species. Explanatory variables included Study Area as a factor, Year as a numerical variable (to take account of linear trends in the response during the five years from which data had been drawn), and an additional binary variable ‘2022’ that separated 2022 data from data drawn from all other years. The Akaike Information Criterion (AIC) was used to compare models with both Year and 2022, and each of these variables separately, to evaluate whether inter-annual differences were best explained by a gradual change through time or by a difference between 2022 and all other years.

2.3 Comparisons with tables from annual reports

The mean proportion of pairs successfully producing young and the mean fledged brood size of successful pairs were compared between 2022 and the data from SRMS Annual Reports covering the years 2019 to 2021. For each species, comparisons were made for all SRMS reporting regions for which 10 or more breeding pairs were reported during 2022, and an average of 10 or more breeding pairs were reported during the previous three years. For each comparison, the rank of the metric in 2022 among its equivalent for other years was recorded. A comparison of Barn Owl breeding success and fledged brood size was made between 2021 and 2022, as this is likely to be a good indicator of short term changes in the availability of small mammal prey (Challis et al., 2022)

2.4 Regional comparisons adjusted for trends

A more detailed analysis was carried out using raw SRMS data for any combination of species and region for which comparisons with data from Annual Reports suggested that breeding success or fledged brood size was lower in 2022 than in any of the three previous years,. Records from the five years from 2018 to 2022 were extracted, and individual GLMs were run for each combination of region (including the whole of Scotland, where this comparison was relevant), species and metric (for breeding success and fledged brood size). Region or area was not included in the models as an explanatory variable (because each model dealt with data from an individual region), but otherwise the model structure was as described for the local studies analysis above.

2.5 Evidence of HPAI impacts from SRMS contributors

Records submitted to the SRMS in 2022 were scrutinised in order to find any comments or evidence recorded by contributors (beyond the occupancy and breeding outcome information we compared with equivalent information from previous years, as described above) that indicated possible impacts of HPAI. An automated search was carried out on the whole 2022 dataset to identify any field containing the character strings "HPAI", "Flu", "bird flu", "H5N1", "postmortem", "post mortem", "avian", and “disease” and "Influenza". Unique values in all fields containing any of these search terms were examined. Any entry mentioning a possible impact of HPAI was categorised according to the type of suggested impact (impact on adult birds, impact on brood, both, or other - the latter including impacts on fieldwork), and robustness of evidence ('Possible' where no supporting observations or diagnoses were provided, 'Probable' where a good case was made but no proof given, and 'Confirmed' where a diagnosis of AI was obtained).

SRMS contributors were also asked, through the Chairs of the Scottish Raptor Study Group regional branches, to submit any extra information they might have relating to impacts of HPAI in 2022 directly to the Coordinator of the SRMS. All such information was examined thoroughly and summarised for this report.

3. Results

3.1 Local Studies

There were sufficient data to compare number of pairs between 2022 and the previous four years for 34 local studies of 10 raptor species (Table 1). Comparisons of breeding success were possible for 24 studies of nine species (Table 2), while comparisons of fledged brood size were possible for 15 studies of nine species (Table 3).

These local study-based comparisons provide scant evidence of impacts of HPAI on the number of pairs recorded in these areas. Accounting for background population change, the number of pairs recorded in 2022 was lower than that for previous years in just three studies: the buzzard study in Tain, Ross-shire (where numbers dropped from an average of 53.2 pairs in 2018 - 2021 to 42 pairs in 2022), the hen harrier study on Rousay, Orkney (where the drop was from 6.3 pairs to two pairs), and the red kite study in Central Scotland (where number of pairs dropped from 58.0 to 45). For the first two of these studies, the number of territories checked in 2022 was slightly higher than in the previous four years; but both of these decreases were only marginally statistically significant. For the red kite study, the number of home ranges reported to have been checked in 2022 was lower than in the previous four years.

Several of the studies summarised in Table 1 actually recorded apparent increases in breeding population size in 2022. These included: golden eagle studies in Mull, Morvern and Ardnamurchan, and in Central Lewis and Harris; the hen harrier study on Mainland Orkney; the red kite study in Dumfries & Galloway; and (with marginal statistical significance) the raven Study on Colonsay. For all of these studies, the number of home ranges recorded as having been checked in 2022 was higher than the average number checked in the previous four years. This means that these apparent increases in breeding population size could have been contributed to by increased survey effort in 2022. It is possible that some of the extra home ranges checked in 2022 were newly established territories but, at least for the golden eagle and raven studies, this seems unlikely, as the increases in 2022 run counter to overall local decreases recorded by these studies from 2018 to 2022.

Local studies provide strong evidence of impacts on breeding success, particularly for the two eagle species. Including significant and non-significant results, 14 out of the 24 studies summarised in Table 2 recorded lower levels of breeding success in 2022 than in the previous four years. Of eight studies for which this difference is statistically significant, seven are of eagles (four for golden eagle and three for white-tailed eagle). The only local eagle study for which breeding success in 2022 was not significantly lower than in previous years is the golden eagle study in Badenoch and West Inverness-shire. The closest active nest site to the coast recorded in this area between 2018 and 2022 was 13 km from the nearest tidal area. By contrast, six of the seven studies where eagles were found to have declined significantly are substantially coastal or (in the case of the Trossachs golden eagle study) contain several recently occupied nest sites within 5 km of sections of the coast and/or sea lochs. The only other study for which breeding success in 2022 was significantly lower than in the previous four years was the raven study on Shetland, for which the number of monitored pairs was higher in 2022 than in the earlier period. Buzzards in Tain, Ross-shire experienced lower breeding success in 2022 than during 2018-2021, but this difference was not statistically significant.

Breeding success for inland white-tailed eagles on mainland Scotland (outside of the largely coastal areas covered by local studies) was actually higher in 2022 than during the previous four years, though this difference was not significant. The osprey studies in Stirlingshire and in Easter Ross & Inverness, the hen harrier study on Mainland Orkney and the red kite study in Central Scotland also recorded higher breeding success in 2022 than in the previous four years: this difference was statistically significant in the Stirlingshire osprey study and marginally significant in the other three.

There is no evidence of any potential impacts of HPAI on 2022 fledged brood sizes from local studies. Of the 15 studies for which there were enough successful breeding attempts recorded to analyse fledged brood size, the only one for which this metric differed (with marginal significance) was the raven study on Shetland, where recorded fledged brood sizes were slightly higher in 2022 than in previous four years.

3.2 Comparisons with tables from annual reports

Table 4 and Table 5 show 2022 breeding success and fledged brood size for region-species combinations represented by sufficient data, and indicate for which of these combinations measures were lower than in any of the previous 3 years.

Kestrel had a relatively poor year in 2022. Breeding success and productivity were not as dramatically depressed in 2022 as these metrics were for eagles, but it is possible that the abundance of voles in many areas helped to counter what might otherwise have been a larger impact of HPAI.

Table 1. Numbers of pairs recorded in local study areas for the relevant species in 2022 and the mean annual number of pairs recorded in the four-year period prior to this. Numbers in brackets are the number (or mean number) of territories checked during these periods. Where the best fitting model included a binary variable representing this period, the species is marked [L] (where the number of pairs was lower in 2022) and [H] (where it was higher). For these studies, the statistical significance of the difference between years is given. Where the best fitting model included a slope representing linear change over this period, the study is labelled ‘Up’ (for increasing trends) or ‘Down’ for decreasing trends.

Species	Study	Pairs 2022	Pairs 2018-21	Significance	Trend
Buzzard [L]	Tain	42 (n=85)	53.2 (n=82.5)	P=0.16	-
Buzzard	Colonsay	9 (n=57)	5.8 (n=47)	NS	-
Buzzard	Bute	10 (n=23)	13.2 (n=48.2)	NS	Down
Golden Eagle	Trossachs	37 (n=44)	30.5 (n=37.8)	NS	Up
Golden Eagle [H]	Mull & Lochaber	33 (n=36)	21 (n=25.2)	P=0.001	Down
Golden Eagle [H]	Central Lewis & Harris	30 (n=31)	13.5 (n=15)	P=0.001	Down
Golden Eagle	NW Sutherland	19 (n=22)	16.2 (n=21.8)	NS	-
Golden Eagle	Badenoch and W Inverness	35 (n=44)	30.5 (n=39)	NS	-
Hen Harrier [H]	Orkney Mainland	77 (n=182)	54.8 (n=163.8)	P=0.01	-
Hen Harrier [L]	Rousay	2 (n=21)	6.3 (n=23)	P=0.121	-
Hen Harrier	Hoy	6 (n=35)	16 (n=37.5)	NS	Down
Kestrel	Orkney Mainland	3 (n=30)	3 (n=28.5)	NS	-
Kestrel	Pentlands	11 (n=16)	12.3 (n=21)	NS	-
Merlin	Orkney Mainland	5 (n=42)	3.2 (n=38.8)	NS	-
Merlin	Shetland	30 (n=48)	28.2 (n=54.8)	NS	-
Merlin	Forest of Birse	8 (n=23)	8.5 (n=25.5)	NS	-
Osprey	Loch Awe	6 (n=13)	5.2 (n=11)	NS	-
Osprey	Stirlingshire	19 (n=37)	22.5 (n=34)	NS	-
Osprey	Easter Ross & Inverness	40 (n=53)	32.5 (n=49.8)	NS	-
Peregrine	Southern Uplands	38 (n=122)	39.5 (n=107)	NS	-
Peregrine	SW Coast	12 (n=22)	10.2 (n=20.2)	NS	-
Peregrine	Lothian & Forth	17 (n=35)	17 (n=35.8)	NS	-
Peregrine	Southeast Coast	13 (n=48)	13 (n=52.8)	NS	-
Peregrine	Dumfries Coast	12 (n=22)	13.5 (n=19.8)	NS	-
Raven	Tiree	5 (n=17)	4 (n=16)	NS	-
Raven [H]	Colonsay	8 (n=23)	6 (n=18.2)	P=0.059	Down
Raven	Shetland	29 (n=67)	18.2 (n=28.8)	NS	Up
Raven	Bute	15 (n=23)	16 (n=25.8)	NS	-
Red Kite [L]	Central Scotland	45 (n=98)	58 (n=112)	P=0.002	Up
Red Kite [H]	Dumfries & Galloway	84 (n=114)	59.8 (n=105.5)	P=0.007	-
W-tailed Eagle	Mull & Lochaber	26 (n=27)	27 (n=27.5)	NS	-
W-tailed Eagle	N Skye	19 (n=25)	15 (n=19.8)	NS	-
W-tailed Eagle	E Lewis	17 (n=17)	11.2 (n=12.2)	NS	-
W-tailed Eagle	INLAND	22 (n=27)	15.8 (n=17.8)	NS	Up

Table 2. Breeding success of pairs in local study areas for the relevant species in 2022 and of pairs in the four-year period prior to this. Numbers in brackets are the number (or mean annual number) of pairs monitored during these periods. Where the best fitting model included a binary variable representing this period, the species is marked [L] (where breeding success was lower in 2022) and [H] (where it was higher). For these studies, the statistical significance of the difference between years is given. Where the best fitting model included a slope representing linear change over this period, the study is labelled ‘Up’ (for increasing trends) or ‘Down’ for decreasing trends.

Species	Study	Success 2022	Success 18-21	Significance	Trend
Buzzard	Tain	0.52 (n=42)	0.59 (n=53.2)	NS	-
Buzzard	Bute	1 (n=10)	0.98 (n=13.2)	NS	Up
Golden Eagle [L]	Trossachs	0.27 (n=37)	0.46 (n=30.5)	P=0.044	-
Golden Eagle [L]	Mull & Lochaber	0.12 (n=33)	0.31 (n=21)	P=0.043	-
Golden Eagle [L]	Central Lewis & Harris	0.17 (n=30)	0.54 (n=13.5)	P=0.002	-
Golden Eagle [L]	NW Sutherland	0.11 (n=19)	0.4 (n=16.2)	P=0.028	-
Golden Eagle	Badenoch & W Inverness	0.43 (n=35)	0.55 (n=30.5)	NS	-
Hen Harrier [H]	Orkney Mainland	0.27 (n=77)	0.18 (n=54.8)	P=0.095	-
Merlin	Shetland	0.67 (n=30)	0.74 (n=28.2)	NS	-
Osprey [H]	Stirlingshire	0.84 (n=19)	0.58 (n=11.2)	P=0.032	Down
Osprey [H]	Easter Ross & Inverness	0.85 (n=40)	0.72 (n=32.5)	P=0.093	-
Peregrine	Southern Uplands	0.61 (n=38)	0.72 (n=39.5)	NS	-
Peregrine	SW Coast	0.5 (n=12)	0.76 (n=10.2)	NS	Down
Peregrine	Lothian & Forth	0.53 (n=17)	0.51 (n=17)	NS	-
Peregrine	SE Coast	0.54 (n=13)	0.62 (n=13)	NS	-
Peregrine	Dumfries Coast	0.92 (n=12)	0.83 (n=13.5)	NS	-
Raven [L]	Shetland	0.55 (n=29)	0.74 (n=18.2)	P=0.022	Up
Raven	Bute	0.87 (n=15)	0.8 (n=16)	NS	-
Red Kite [H]	Central Scotland	0.84 (n=45)	0.72 (n=58)	P=0.078	-
Red Kite	Dumfries & Galloway	0.87 (n=84)	0.82 (n=59.8)	NS	Up
W-tailed Eagle [L]	Mull & Lochaber	0.5 (n=26)	0.67 (n=27)	P=0.117	-
W-tailed Eagle [L]	North Skye	0.16 (n=19)	0.6 (n=15)	P=0.002	-
W-tailed Eagle [L]	E Lewis	0.35 (n=17)	0.64 (n=11.2)	P=0.015	Up
W-tailed Eagle	INLAND	0.77 (n=22)	0.67 (n=15.8)	NS	Up

Table 3. Fledged Brood size of successfully breeding pairs in local study areas for the relevant species in 2022 and of successful pairs in the four-year period prior to this. Numbers in brackets are the number (or mean annual number) of successful pairs monitored during these periods. Where the best fitting model included a binary variable representing this period, the species is marked [L] (where breeding success was lower in 2022) and [H] (where it was higher). For these studies, the statistical significance of the difference between years is given. Where the best fitting model included a slope representing linear change over this period, the study is labelled ‘Up’ (for increasing trends) or ‘Down’ for decreasing trends.

Species	Study	Fledged 2022	Fledged 18-21	Significance	Trend
Buzzard	Tain	1.82 (n=22)	1.67 (n=31.2)	NS	-
Buzzard	Bute	1.3 (n=10)	1.54 (n=13)	NS	-
Golden Eagle	Trossachs	1 (n=10)	1.05 (n=14)	NS	-
Golden Eagle	Badenoch & W Inverness	1.2 (n=15)	1.4 (n=16.8)	NS	-
Hen Harrier	Orkney Mainland	2.05 (n=21)	2.02 (n=10)	NS	-
Merlin	Shetland	3.55 (n=20)	3.52 (n=21)	NS	-
Osprey	Easter Ross & Inverness	1.88 (n=34)	1.89 (n=23.2)	NS	-
Peregrine	Southern Uplands	2.61 (n=23)	2.35 (n=28.2)	NS	-
Peregrine	Dumfries Coast	1.82 (n=11)	1.6 (n=11.2)	NS	-
Raven [L]	Shetland	3.06 (n=16)	3.02 (n=13.5)	P=0.072	Down
Raven	Bute	2.23 (n=13)	2.67 (n=12.8)	NS	-
Red Kite	Central Scotland	1.47 (n=48)	1.75 (n=41.5)	NS	-
Red Kite	Dumfries & Galloway	1.33 (n=73)	1.24 (n=49)	NS	-
W-tailed Eagle	Mull & Lochaber	1.15 (n=13)	1.18 (n=18)	NS	-
W-tailed Eagle	INLAND	1.18 (n=17)	1.26 (n=10.5)	NS	-

Table 4. Breeding success of monitored pairs by SRMS region in 2022 for raptor species with 10 or more monitored pairs in that region. Numbers in brackets are the number (or mean annual number) of pairs monitored in 2022. Where the breeding success in 2022 is lower than in any of the previous three years, the species is marked [L]. Individual species tables giving the mean breeding success and sample sizes in each region for each year from 2019 to 2022 are available on request.

Region	Buzzard	Golden Eagle	Goshawk	Hen Harrier	Kestrel	Merlin	Osprey	Peregrine	Raven	Red Kite	Sparrow-hawk	W-tailed Eagle
Argyll	0.81 (31)	0.27 (48) [L]	-	0.83 (18)	-	-	0.7 (10)	0.89 (9)	0.82 (55)	-	-	0.55 (33) [L]
Central	-	-	-	-	0.88 (42) [L]	-	0.73 (30)	0.64 (11) [L]	0.81 (21)	0.77 (22)	0.92 (12)	-
D&G	0.68 (22)	-	0.71 (21)	-	-	-	0.82 (11)	0.66 (44)	0.87 (23)	0.85 (125)	-	-
Highland	0.64 (84)	0.31 (100) [L]		0.79 (24)	0.94 (17)	0.67 (18) [L]	0.82 (66)	0.75 (28)	0.89 (27) [L]	0.64 (39) [L]	0.94 (17)	0.45 (60) [L]
Lewis & Harris	-	0.16 (31) [L]	-	0.8 (15)	-	-	-	-	-	-	-	0.24 (25) [L]
Lothian & Borders	0.81 (88)	-	0.71 (35)	-	0.65 (26)	0.89 (9)	0.85 (13)	0.5 (48) [L]	0.89 (35)	-	-	-
NE Scotland	-	-	-	-	-	0.93 (28)	0.75 (20)	0.88 (17)	-	0.75 (20) [L]	-	-
Orkney	0.58 (12)	-	-	0.28 (89)	0.64 (14)	0.6 (10) [L]	-	0.4 (10)	0.45 (29) [L]	-	0.53 (19)	-
Shetland	-	-	-	-	-	0.69 (39)	-	-	0.55 (29) [L]	-	0.65 (23) [L]	-
S Strathclyde	-	-	-	-	-	-	-	0.85 (33)	0.87 (23)	-	-	-
Tayside	0.86 (66) [L]	0.5 (24) [L]	-	0.64 (11)	0.7 (10) [L]	-	-	0.58 (31) [L]	0.85 (40)	0.90 (42)	-	-
Uist	0.67 (9)	-	-	0.4 (15) [L]	-	-	0.94 (32)	-	0.65 (17)	-	-	-
TOTAL	0.76 (319) [L]	0.3 (223) [L]	0.76 (88)	0.51 (196)	0.78 (130) [L]	0.77 (121)	0.82 (185)	0.67 (234) [L]	0.77 (300) [L]	0.8 (203)	0.77 (102)	0.45 (130) [L]

Table 5. Fledged brood size of successful pairs by SRMS region in 2022 for raptor species with 10 or more successful pairs were monitored in that region. Numbers in brackets are the number (or mean annual number) of successful pairs monitored in 2022. Where fledged brood size in 2022 is lower than in any of the previous three years, the species is marked [L]. Individual species tables giving the mean breeding success and sample sizes in each region for each year from 2019 to 2022 are available on request.

Region	Buzzard	Golden Eagle	Goshawk	Hen Harrier	Kestrel	Merlin	Osprey	Peregrine	Raven	Red Kite	Sparrow-hawk	W-tailed Eagle
Argyll	1.36 (25) [L]	1 (13) [L]	-	3.72 (15)	-	-	1.57 (7)	-	2.81 (45)	-	-	1.1 (18) [L]
Central	-	-	-	-	3.97 (37)	-	2.18 (22)	-	2.22 (17)	1.12 (17) [L]	3.16 (11)	-
D&G	1.76 (15)	-	1.96 (15)	-	-	-	-	2.28 (29)	1.27 (20) [L]	1.3 (106)	-	-
Highland	1.56 (54)	1.1 (31) [L]	-	2.02 (19) [L]	3.4 (16)	2.7 (12)	1.96 (54)	1.87 (21)	3.26 (24)	1.87 (25)	2.44 (16)	1.11 (27)
Lewis & Harris		-	-	2.63 (12)	-	-	-	-	-	-	-	-
Lothian & Borders	0.87 (71) [L]	-	-	-	3.82 (17)	-	2.6 (11)	2.6 (24) [L]	2.37 (31) [L]	-	-	-
NE Scotland	-	-	-	-	-	2.37 (26) [L]	2.67 (15)	1.7 (15)	-	2 (15)		-
Orkney	-	-	-	2.14 (25)	-	-	-	-	2.9 (13)	-	2.66 (10)	-
Shetland	-	-	-	-	-	3.32 (27) [L]	-	-	3.08 (16)	-	2.3 (15)	-
S Strathclyde	-	-	-	-	-	-	1.81 (30)	2.24 (28)	2.88 (20)	-	-	-
Tayside	1.39 (57)	1.08 (12) [L]	-	-	-	-	-	2.41 (18)	2.47 (34)	1.61 (38) [L]	-	-
Uist	-	-	-	-	-	-	-	-	2.01 (11)	-	-	-
TOTAL	1.58 (242)	1.06 (67) [L]	1.71 (67)	2.77 (99)	3.6 (101)	2.73 (93) [L]	2.08 (151)	2.24 (157)	2.72 (232)	1.5 (162) [L]	2.58 (79)	1.12 (58) [L]

Merlins feed mainly on small birds and appeared to have lower breeding success in 2022 than in previous years in two out of the three regions for which there were enough data to robustly assess this (though this did not translate to lower breeding in 2022 across the whole Scottish dataset). Merlin fledged brood sizes in 2022 were lower than in previous years in two out of the three regions with sufficient sample sizes, as well as at a national level.

Across the whole of Scotland, for local studies from which we have available data, breeding performance of buzzard and red kite populations in 2022 appears to have been similar to 2021 levels of breeding success and fledged brood size. This could indicate relatively little impact of HPAI, although it is possible that some negative impacts were counteracted by other factors such as favourable weather or improved food availability.

Breeding success of both raven and peregrine in 2022 was lower than in the previous three years for three regions (Central, Lothian & Borders and Tayside for peregrine; and Highland, Orkney and Shetland for raven) out of a total of 10 for which sample sizes were sufficient for peregrine, and 11 for raven. The overall breeding success of these species for the whole Scotland-wide dataset in 2022 was also lower than in previous years. Fledged brood sizes for these species in 2022 were lower than in previous years for only three regions across both species: Lothian & Borders for peregrine, and Lothian & Borders and Dumfries & Galloway for raven.

White-tailed eagle and golden eagle populations were much less productive in 2022 than in previous years. In all regions for which sample sizes were sufficient (five for golden eagle and three for white-tailed eagle), as well as across all pairs monitored in Scotland, breeding success in 2022 was lower than in any of the previous four years. Impacts on fledged brood size were also evident, despite the small clutch and brood sizes of these species (meaning that there is limited scope for variation between years). Pairs that were successful fledged smaller broods in all three regions for which golden eagle sample sizes were sufficient, in one of the two regions for which white-tailed eagle sample sizes were sufficient, and at a Scotland-wide level for both species.

Measures of productivity for osprey, goshawk, and sparrowhawk in 2022 appeared relatively unaffected in most regions. Out of a total of 24 regional measures of breeding success and fledged brood size with sample sizes of 10 or more for these species, the only one that was lower than the equivalent measure in the previous three years was for sparrowhawk in Shetland.

Barn owl breeding productivity was substantially higher year in 2022 than in 2021. In all regions with sufficient sample sizes to allow comparisons between these years, both breeding success and fledged brood size were higher. Across the whole SRMS dataset, barn owl breeding success increased from 0.63 in 2021 to 0.84 in 2022, while mean fledged brood size rose from 2.21 young fledged per successful pair to 2.85.

3.3 Regional comparisons adjusted for trends

Breeding success (Table 6) was significantly lower in 2022 than in previous years for many of the species-region combinations for which annual table comparisons suggested a difference between years. Breeding success of both eagle species in 2022 was lower in all seven regions where sample sizes allowed a comparison with previous years. Golden eagle breeding success in 2022 was 0.14 – 0.39 lower than in previous years, while white-tailed eagle breeding success was 0.1 - 0.42 lower.

There were also several statistically significant but widely scattered regional declines in breeding success among other species, including hen harrier (Uist), kestrel (Central Scotland), Peregrine (Tayside & Fife), raven (both Orkney and Shetland) and red kite (both North East Scotland and Highland). These declines varied in severity from 0.08 for kestrel in Central Scotland to 0.49 for hen harrier in Uist. There were three significant increases in breeding success among these species-region combinations: for merlin in North East Scotland, and for red kite in Central Scotland and Tayside.

Regional comparisons of fledged brood size (Table 7) suggest smaller negative declines of this aspect of breeding productivity in 2022. Negative changes between 2022 and previous years were statistically significant for just two species-region combinations; merlin in Shetland and raven in Dumfries & Galloway. There were also statistically significant increases in fledged brood size for these species elsewhere: in Highland for merlin and in both Highland and Shetland for raven. Fledged brood sizes recorded in 2022 for both eagle species in all five regions with 10 or more successful pairs were smaller than in previous years, but these differences were not statistically significant.

Table 6. Breeding success of pairs in SRMS regions for which comparisons with tables from SRMS annual reports indicated that either breeding success or fledged brood size were lower in 2022 than in previous years (Table 2). Mean breeding success is given for 2022 and for the 4 years from 2018 to 2021. Numbers in brackets are the number (or mean annual number) of pairs monitored during these periods. Where the best fitting model included a binary variable representing this period, the species is marked [L] (where breeding success was lower in 2022) and [H] (where it was higher). For these studies, the statistical significance of the difference between years is given. Where the best fitting model included a slope representing linear change over this period, the study is labelled ‘Up’ (for increasing trends) or ‘Down’ for decreasing trends.

Species	Region	Success 2022	Success 2018-21	Significance	Trend
Buzzard	Tayside	0.86(n=66)	0.92(n=132)	NS	Down
Buzzard	SCOTLAND	0.83(n=185)	0.89(n=240)	NS	Down
Buzzard	Argyll	0.81(n=31)	0.85(n=39.5)	NS	Down
Buzzard	Lothian & Borders	0.81(n=88)	0.84(n=68.5)	NS	-
Golden Eagle [L]	Argyll	0.26(n=66)	0.4(n=53.8)	P = 0.038	-
Golden Eagle [L]	Highland	0.29(n=121)	0.49(n=92.5)	P < 0.001	-
Golden Eagle [L]	Lewis & Harris	0.16(n=31)	0.55(n=19.2)	P = 0.001	-
Golden Eagle	Tayside	0.48(n=25)	0.58(n=19.5)	NS	-
Golden Eagle [L]	SCOTLAND	0.28(n=243)	0.48(n=185)	P < 0.001	-
Hen Harrier [L]	Uist	0.4(n=15)	0.89(n=13.5)	P < 0.001	-
Hen Harrier	Highland	0.79(n=24)	0.62(n=25.2)	NS	Up
Kestrel [L]	Central	0.88(n=42)	0.96(n=22.5)	P = 0.127	-
Kestrel	Tayside	0.7(n=10)	0.91(n=22)	NS	Down
Kestrel [L]	SCOTLAND	0.85(n=52)	0.93(n=44.5)	P = 0.058	-
Merlin	Highland	0.67(n=19)	0.79(n=28.5)	NS	Up
Merlin [H]	NE Scotland	0.9(n=31)	0.76(n=34)	P = 0.088	Up
Merlin	Shetland	0.56(n=32)	0.72(n=39.2)	NS	Up
Peregrine	Central	0.72(n=18)	0.82(n=13.8)	NS	-
Peregrine	Lothian & Borders	0.5(n=48)	0.58(n=49.2)	NS	-
Peregrine [L]	Tayside	0.56(n=32)	0.72(n=39.2)	P = 0.013	Up
Peregrine [L]	SCOTLAND	0.56(n=98)	0.67(n=102.2)	P = 0.015	Up
Raven	D&G	8.9(n=27)	9.6(n=28.2)	NS	-
Raven	Highland	0.89(n=27)	0.96(n=28.2)	NS	-
Raven [L]	Orkney	0.45(n=29)	0.63(n=21.5)	P = 0.093	-
Raven [L]	Shetland	0.53(n=30)	0.72(n=17.8)	P = 0.025	Up
Raven [L]	SCOTLAND	0.62(n=86)	0.79(n=67.5)	P = 0.002	-
Red Kite [H]	Central	0.77(n=22)	0.61(n=26.8)	P = 0.149	-
Red Kite [L]	Highland	0.64(n=39)	0.84(n=35)	P = 0.007	-
Red Kite [L]	NE Scotland	0.75(n=20)	0.85(n=15.5)	P = 0.008	Up
Red Kite [H]	Tayside	0.90(n=42)	0.80(n=49.0)	P = 0.037	Up
Red Kite	SCOTLAND	0.77(n=123)	0.78(n=126.2)	NS	-
Sparrowhawk	Orkney	0.53(n=19)	0.44(n=13.5)	NS	Up
W-tailed Eagle [L]	Argyll	0.58(n=36)	0.68(n=35.5)	P = 0.028	Up
W-tailed Eagle [L]	Highland	0.46(n=68)	0.64(n=49.8)	P = 0.009	-
W-tailed Eagle [L]	Lewis & Harris	0.24(n=25)	0.66(n=18.5)	P < 0.001	Up
W-tailed Eagle [L]	SCOTLAND	0.45(n=135)	0.67(n=112.5)	P < 0.001	-

Table 7. Fledged brood size of successful pairs in SRMS regions for which comparisons with tables from SRMS annual reports indicated that either breeding success or fledged brood size were lower in 2022 than in previous years (Table 3). Mean Fledged brood sizes are given for 2022 and for the 4 years from 2018 to 2021. Numbers in brackets are the number (or mean annual number) of pairs monitored during these periods. Where the best fitting model included a binary variable representing this period, the species is marked [L] (where breeding success was lower in 2022) and [H] (where it was higher). For these studies, the statistical significance of the difference between years is given. Where the best fitting model included a slope representing linear change over this period, the study is labelled ‘Up’ (for increasing trends) or ‘Down’ for decreasing trends.

Species	Region	Fledge 2022	Fledge 2018-21	Significance	Trend
Buzzard	Tayside	1.46(n=57)	1.35(n=121.2)	NS	-
Buzzard	SCOTLAND	1.38(n=153)	1.5(n=212.5)	NS	Down
Buzzard	Argyll	1.36(n=25)	1.56(n=33.8)	NS	-
Buzzard	Lothian & Borders	1.32(n=71)	1.8(n=57.5)	NS	Down
Golden Eagle	Argyll	1(n=17)	1.02(n=21.5)	NS	-
Golden Eagle	Highland	1.09(n=35)	1.19(n=45)	NS	-
Golden Eagle	Tayside	1.08(n=12)	1.33(n=11.2)	NS	-
Golden Eagle	SCOTLAND	1.06(n=69)	1.15(n=88.2)	NS	-
Hen Harrier	Highland	2(n=19)	2.65(n=15.8)	NS	Down
Kestrel	Central	4.03(n=37)	3.95(n=21.5)	NS	-
Kestrel	SCOTLAND	3.93(n=44)	3.19(n=41.5)	NS	Up
Merlin [H]	Highland	2.67(n=12)	2.42(n=22.8)	P = 0.02	Down
Merlin [L]	Shetland	3.37(n=27)	3.51(n=25.8)	P = 0.02	Down
Peregrine	Central	2.31(n=13)	2.42(n=11.2)	NS	-
Peregrine	Lothian & Borders	2.67(n=24)	2.82(n=28.5)	NS	-
Peregrine	Tayside	2.33(n=18)	2.02(n=28.2)	NS	-
Peregrine	SCOTLAND	2.47(n=55)	2.42(n=68)	NS	-
Raven [L]	D&G	1.50(n=20)	2.20(n=24.5)	P = 0.001	Down
Raven [H]	Highland	3.21(n=24)	2.44(n=27)	P = 0.001	Down
Raven	Orkney	2.92(n=13)	2.93(n=13.5)	NS	-
Raven [H]	Zetland	3.12(n=16)	3.06(n=12.8)	P = 0.108	Down
Raven [H]	SCOTLAND	3.11(n=53)	2.71(n=53.2)	P = 0.006	Down
Red Kite	Central	1.12(n=17)	1.57(n=16.2)	NS	-
Red Kite	Highland	1.92(n=25)	1.85(n=29.5)	NS	-
Red Kite	North East Scotland	2(n=15)	1.89(n=13.2)	NS	-
Red Kite	Tayside	1.61(n=38)	1.83(n=39.2)	NS	-
Red Kite	SCOTLAND	1.95(n=40)	1.86(n=42.8)	NS	-
W-tailed Eagle	Argyll	1.1(n=21)	1.23(n=24.2)	NS	-

3.4 Evidence of HPAI impacts from SRMS contributors

Out of 5883 raptor records submitted to the SRMS for the breeding season of 2022, a total of 21 reported suspected (17) or definite (four) cases of HPAI.

The four definite cases included one golden eagle nestling (see information reported directly by Derek Spencer, below), three white-tailed eagle nestlings from two separate nests on Mull and Skye, and a pair of adult buzzards from a site north of Perth. HPAI was confirmed by laboratory test at the three eagle sites, and although testing was not mentioned in the buzzard record, the pair were ‘handed in to Police’ and described as definite Bird Flu victims.

The seventeen suspected cases involved six instances of dead young being found at or near the nest site (one raven in Shetland, one buzzard in Fife and four white-tailed eagles in Argyll and Sutherland), five instances of dead adults found at or near the nest (three red kites in Ross-shire and Dumfries & Galloway, and two ravens in Badenoch and Shetland), and four abandoned breeding attempts (three Ross-shire red kites and one white-tailed eagle on Skye).

Among records where observers did not mention HPAI, a further one record reported a dead adult at the nest (a barn owl in Dumfries & Galloway) and 11 records reported dead young in circumstances that did not obviously point to predation as a cause of death (these involved five hen harrier nesting attempts in Orkney, four barn owl sites in Dumfries & Galloway, one barn owl site in Ross-shire, and one white-tailed eagle nest on Rum).

In addition to the information contained in 2022 SRMS records, three SRMS contributors responded to the request for any additional information on evidence for impacts of HPAI. All three respondents are members of Highland branch of the Scottish Raptor Study Group (HRSG).

Keith Duncan, Chair of HRSG, provided information from a relatively new study of buzzards which commenced in 2020, and where both the number of breeding pairs appeared to be greatly reduced in 2022 (when seven nests were found, with five of these successful) compared to the previous two years (when 13 and 11 nests were found, with 11 successful in each year). In 2022, most of the vacant buzzard home ranges within this study area (which contains 14 buzzard home ranges in total) were on the west side of this area, adjacent to an area of the Spey floodplain that supports wintering and spring passage flocks of greylag and pink-footed geese.

Similarly, Bob Swann who monitors the long-running local study of buzzards in Tain, responded to say that several regularly occupied red kite and buzzard territories in the area he monitors were occupied only by single birds, or were suddenly abandoned during nesting attempts in 2022. Dead adults (one buzzard and two red kites) were found below active nests in this area. Bob attempted to get the carcasses of these birds picked up by the police (as per existing guidelines) to be tested, but without any success. As with Keith, Bob thought these impacts were related to the spring passage of pink-footed geese – he observes that dead and dying geese were scattered over foraging fields and roost sites in this area, and that these were then scavenged by raptors.

Finally, Derek Spencer described finding a dead golden eagle chick beneath its eyrie in Sutherland. This was the only eaglet in the nest, so the incident comprised a failed breeding attempt. The carcass was picked up by police and tested positive for HPAI.

All of the information provided by these three observers pertaining to individual breeding attempts was also submitted by them and extracted independently from 2022 records. The full responses of these observers, along with supporting photos and information, are available on request.

4. Discussion

4.1 Species affected

The breeding success of golden eagles and white-tailed eagles appears to have been severely impacted in 2022. Across the whole SRMS dataset, golden eagle breeding success declined from 0.48 to 0.28. White-tailed eagle breeding success declined from 0.67 to 0.45. The largest declines recorded for both species were in Lewis & Harris, where breeding success of golden eagles fell from 0.55 to 0.16, and breeding success of white-tailed eagle declined from 0.66 to 0.24.The fact that these species are both scavengers and predators of larger bird species, including water birds whose populations are known to have been impacted by HPAI in 2022, has been recognised as putting these species at higher risk of infection than most other UK raptors (Veen et al., 2007).

Among other raptor species, transmission from other species would be more likely to occur through either one of two routes (via live prey or scavenged carrion). Peregrine, merlin, sparrowhawk and goshawk would be much more likely to contract HPAI from live prey than from scavenged carcasses. Of these species, the size range of prey and coastal distribution of peregrine is likely to put them at greater risk than the other avian predators. Buzzard, raven and red kite are all regular scavengers, and in most situations do not specialise on hunting birds (though some red kites and buzzards regularly hunt birds which could themselves be susceptible to contracting HPAI, such as corvids and pigeons). All of these latter species might be at risk of coming into contact with dead and dying HPAI victims in areas where there were outbreaks in populations of medium and large birds (e.g. seabirds or water birds). Breeding success in 2022 of peregrines, ravens and buzzards was lower across the whole of Scotland, with evidence of local and regional impacts being stronger (and supported by observer evidence) for the latter two.

A superficial comparison between these findings and recent confirmed cases of HPAI in raptors (Animal and Plant Health Agency, 2022) might suggest markedly different patterns. The overwhelming majority of raptors that have tested positive for HPAI in Britain are buzzards (132), with numbers among other species being much lower (14 sparrowhawks, six red kites, six white-tailed eagles and two golden eagles). However, when interpreted in the light of UK breeding population sizes (Woodward et al., 2021), these findings are more consistent with the conclusions of this report. The ratio of HPAI positive buzzard carcasses to breeding pairs in Britain is approximately 1:500 breeding pairs, with the equivalent ratio for several other species being considerably higher – notably 1:20 for white-tailed eagle and 1:250 for both peregrine and golden eagle. There are also several other factors that may confound and complicate a relationship between the numbers of carcasses reported as HPAI positive and impacts of HPAI on breeding raptors. Effective search effort and encounter rates for raptor carcasses are likely to vary markedly between different areas, such that the proportion of HPAI-related deaths detected for species found mostly in remote areas is likely to be lower than for raptors commonly found in areas of high human population density. Also, non-fatal HPAI infections cannot be detected by carcass testing, but could still impact on breeding productivity if sub-lethal effects of HPAI render breeding adults unable to produce or hatch eggs or successfully rear offspring.

Relative to threats of transmission from prey and carrion, the risk of lateral transmission among individuals of the same species is likely to be lower for raptors than for other groups such as seabirds and water birds, due to the fact that most species are not gregarious. However, there are some notable exceptions. Non-breeding ravens can form large flocks in some areas (Wilson et al., 2019), and wintering harriers and owls often roost communally (Hardey et al. 2013). Red kites and white-tailed eagles are both known to roost communally in winter with the roosts of the former species in Scotland sometimes reaching sizes of 50 or more individuals (Hardey et al., 2013). Regardless of roost size, however, it is not clear whether birds within a typical raptor roost would have sufficient contact with one another to result in substantial risk of disease transmission. A coordinated count at Scottish roosts of red kite in January of 2023 was higher than the equivalent count in 2022, suggesting that no major decline in the intervening period. Perhaps a greater risk than communal roosts is posed by interactions between scavenging species such as ravens, buzzards, red kites and eagles at carcasses, as these can result in close contact between individuals of the same or of different species.

4.2 Areas affected

From region-based comparisons of productivity measures between 2022 and previous years, differences between species appear to be larger and more consistent than those between different parts of Scotland. However, among populations of both of the species for which breeding success in 2022 was strikingly low, golden eagle and white-tailed eagle, apparent impacts were more severe in coastal areas than in more inland populations. As mentioned above, both of these species are known to predate upon seabirds in coastal areas. During late autumn, winter and early spring, eagles in many coastal areas, especially in regions that are on the flyways of migratory geese (such as Islay and the Outer Hebrides) may take a lot of geese and other wildfowl. As both seabirds and wildfowl are the two groups among which incidence and impacts of HPAI appear to be highest, there seems likely to be high potential for inter-species transmission of disease to eagles along the north-west coastlines of mainland Scotland and the island groups inhabited by breeding eagles.

From the comparisons of data between 2022 and previous years in study areas and SRMS regions, as well as from the evidence submitted independently by members of HRSG, there appear to have been localised impacts on other species, particularly on scavenging species such as ravens in Orkney and Shetland and buzzards and red kites in parts of eastern Highlands. None of these are so striking, however, that there can be any certainty that they are due to impacts of HPAI rather than to other factors such as variation in survey effort (see below), or local impacts of food availability or weather. However, these observations are consistent with localised opportunities for scavenging on carcasses of HPAI victims of species in which outbreaks are known to have occurred in 2022, including colonially breeding seabirds and wintering and passage flocks of geese. The possibility of transmission to raptors from these different sources suggests that HPAI outbreaks could remain a threat to resident raptor populations year round.

4.3 Other factors that could explain observed effects

Weather can make it hard for breeding adults to find sufficient food to sustain them and their broods (Newton 1979). Much of 2022 was unusually dry and warm, with many months being among the warmest on record. A cold snap with snowfalls at the start of April (Tanno, 2022), following an exceptionally mild March, could have negatively affected breeding raptors, especially early-breeding species. Raven, golden eagle and white-tailed eagle, three of the species for which breeding success appeared to be most negatively impacted, are all very early breeders. Many of pairs of these species would have been on eggs during the 2022 cold snap, making them vulnerable to weather-related impacts. However, comparisons of monthly minimum temperatures and frost days for Scotland suggest that conditions during March, April and May of 2022 were not markedly colder than those in previous years, with April conditions in 2021 being substantially more severe (MetOffice, 2023). Moreover, while the effects of cold weather and snowfalls on raptors are typically more extreme inland (Ratcliffe, 1993), the areas where breeding success of these early-breeding species seemed most depressed in 2022 were coastal. Thus, while it is possible that the cold snap in 2022 had a negative effect on breeding success of some raptors, it is unlikely to be the main driver of the prominent differences in breeding success between 2022 and previous years.

Unlike for temperatures, rainfall is often more extreme in coastal than for inland areas. 2022 rainfall in the Met Office regions of Northern and Western Scotland was higher than average in the months of February (highest since 2020 in both regions), April (highest since 2013 in Northern Scotland and since 2018 in Western Scotland) and May (highest since 2015 in both regions). Rain can impact negatively on breeding success and nestling survival of raptors by reducing survival of unattended chicks and, over longer periods, reducing adult hunting success, which can result in reduced ability to provision chicks and also (by impacting on adult condition) to keep eggs and chicks warm and dry (Newton, 1979). Higher levels of rainfall during pre-laying, laying, incubation and chick-rearing periods could have negatively impacted on the breeding success of eagles in 2022. However, like temperature, rainfall in 2022 was not so extreme that is seems likely it is the sole factor responsible for lower breeding success in 2022 than in previous years.

Variation in abundance of prey and also breeding productivity of prey populations (because of the importance of nestlings and recently fledged juveniles as prey items for many raptors at the time of year when they are provisioning their own nestlings and fledglings) can also impact on raptor breeding performance (Newton, 1979). One of the most robust indicators of annual variation in the population status and breeding performance of avian prey are the indices of abundance change and demographic measures derived from schemes like the BTO/JNCC/RSPB Breeding Bird Survey (BBS) and Nest Record Scheme (NRS), published in the BBS Annual Report (e.g. Harris et al., 2022) or available online (e.g. BTO 2023). However, metrics from 2022 data gathered by these schemes have not yet been published. An alternative way in which annual variation in abundance of prey may be assessed is through responses of predator populations. This can be particularly effective for assessing fluctuations in the abundance of small mammal prey, for which there are no central sources of information that can be used to make robust comparisons of abundance between years or regions. Barn Owl breeding numbers, success and fledged brood size can all respond to variation in vole abundance (e.g. Challis et al., 2022).

4.4 Representativeness

The recently published SRMS population trends report (Wilson et al., 2022) describes the representativeness of sampling for SRMS data at regional and national levels. There are several combinations of regions and species for which sampling is sufficiently thorough and representative to produce robust trends in breeding productivity. However, white-tailed eagle is the only species for which both number and breeding productivity trends can be considered to represent changes across the whole of Scotland (though recent increases in the number of white-tailed eagles breeding outside of well-monitored areas suggest that the proportion of the Scottish population that is represented by this trend is decreasing). The only other trends that can be taken to represent changes at the whole-Scotland level are productivity trends for golden eagle and osprey. Especially among common raptors like buzzards and kestrels, the samples of territories monitored cannot be relied on to be representative of even regional populations.

It is likely that, for most species considered in this report, any impacts of HPAI that were widespread within the national population will have been detected. Within local studies, too, comparisons of data from previous years are likely to be reliable in picking up impacts. However, at the regional level, our findings may not provide a full reflection of HPAI impacts. Any apparent lack of impacts may belie effects of HPAI outside of the areas and habitats from which SRMS data are drawn. In summary, if all or most individuals of any particular species were impacted by HPAI, we can be confident that the sample data from the SRMS would have detected substantial impacts. However, where an impact of HPAI was detected in a local study, it cannot be inferred that the impact is occurring across a whole region (or nationally) from the sample data available.

By the same token, reduced levels of breeding productivity consistent with the effects of HPAI at a regional level may be very localised and should not be interpreted as occurring across a whole region without further supporting evidence.

4.5 Impacts on survey effort

In some areas, HPAI may have impacted field effort due to organisational restrictions to ringing and bird handling activities, or to concerns of individual raptor workers about the risks posed by nest visits to themselves or to raptor broods. Similarly, movement restrictions imposed by government during the recent COVID-19 pandemic may have negatively impacted on field effort for one of the four years from which data to compare with 2022 breeding records were drawn. Such impacts are likely to have been minimal following well-publicised exemptions to these restrictions for raptor workers were agreed by Scottish government during the summer of 2020, due to the very low risk of COVID-19 transmission during raptor fieldwork. Impacts are therefore likely to have been restricted to the spring of 2020, during which some species-region combinations may have had smaller numbers of pairs being reported and a higher proportion of early-failing pairs missed (resulting in higher reported levels of breeding success) than in other years.

Reduced field effort, particularly early in the breeding season, is likely to result in fewer territorial pairs being recorded. Even when the number of territories checked is unaffected, reductions in the number of (especially early) visits to a territory can reduce the chances of (especially unsuccessful) breeding pairs being detected. Reduced survey effort can also affect measures of productivity. Where reductions in the number of survey visits early in the season lead to some breeding attempts going undetected, most of these are likely to be early failing pairs, as successful breeding attempts are more likely to be picked up on later visits. Field effort early in the season can therefore negatively impact on recorded breeding success. Reduced effort can mean that pairs that fail early go undetected, resulting in a higher proportion of detected pairs going on to breed successfully. This may have been the case with red kites in Central Scotland and Tayside in 2022. There were significantly fewer pairs reported from the Central Scotland local study area, which overlaps with both of these regions. The fact that fewer home ranges were checked in this region suggests that this drop in pair numbers was due to decreased survey effort in 2022. This may have contributed to the increased breeding success of birds in both of these regions, through pairs that failed early on going undetected.

Several, but not all, of the regions where declines in breeding success were recorded, particularly for eagles, experienced modest increases in survey effort and in two study areas (Central Lewis & Harris and Mull & Lochaber) the number of pairs recorded was more than in previous years. As explained above, increased detection of early-failing pairs can lead to a drop in recorded levels of breeding success. In previous national survey years for golden eagle, increased effort has had this effect, increasing the number of pairs found and decreasing the mean level of success recorded across all monitored pairs (Hayhow et al., 2017). However, increases in the number of pairs found in 2022 are not large enough to explain the declines in breeding success. For example, even in the unlikely event that all of the extra pairs of golden eagles found across Scotland in 2022 were early-failing pairs, if the remaining pairs had unchanged breeding success from previous years this would have resulted in an overall breeding success rate of 0.40 in 2022 – substantially higher than the level of 0.28 observed.

4.6 Contributor observations

When considering the strength of evidence reported directly by SRMS data contributors, it is important to bear in mind that, although we asked contributors to include any observations relevant to HPAI from the 2022 field season, neither the structure of the information routinely reported to the SRMS, nor the fieldwork protocols of the majority of fieldworkers who assemble and report records to the Scheme (most of whom follow the guidance in Hardey et al., 2013), are designed explicitly to detect or document the effects of disease.

In addition to the information on territorial occupancy and breeding outcomes, the anecdotal observations contributed by raptor recorders suggest that many breeding failures in 2022 were caused by HPAI. In general, these observations fit well with the observed variation in breeding success and other metrics between species and regions, including records of breeding failures in pairs of golden eagle, white-tailed eagle, buzzard, red kite and raven. Among these observations are descriptions of dead or ill young or adults, suggesting that infection could affect breeding success directly by causing the death of infected nestlings, as well as indirectly by causing the death of one or both parents before the young are fully independent. Where descriptions include mention of laboratory tests, the results of these tests were mostly positive for HPAI, suggesting that the interpretations of many data contributors who suspected impacts of HPAI, even in circumstances where no testing was done, may often be credible. Mortality and sickness of adults and chicks due to other factors such as second generation anticoagulant rodenticides (SGARs) cannot be ruled out, and SGARs have impacted on red kites in previous years (Sansom et al., 2016). However, there is no obvious reason why the impact of such factors would have been higher in 2022 than in other years.

4.7 Further information needs

The evaluation of evidence for possible impacts of HPAI from SRMS data should be repeated following collation of monitoring data from 2023, in order to assess whether impacts of HPAI are still apparent, and whether any of these impacts appear to be more or less severe than in 2022. As well as determining whether HPAI still seems to be depressing breeding success of the two eagle species, both of which are of particularly high conservation importance, it would be useful to improve our understanding of the extent and severity of impacts on other species, for which impacts appeared to be weaker or less consistent in 2022 than those on eagles. Monitoring effort to record territorial occupancy and breeding success should be maintained and enhanced, ensuring that, wherever possible, timing and number of survey visits are sufficient to detect early failing pairs and minimise the frequency of unknown breeding outcomes.

Currently the stage of life cycle at which HPAI is most impacting on breeding raptors is not known, and in future it will be helpful to ensure that SRMS survey visits are frequent enough to discern the stage at which unsuccessful attempts fail (that at least the four standard visits usually recommended in Hardey et al. (2013) are carried out). This may yield useful information that could be considered by future reports on HPAI impacts on raptors. We could not formally assess differences in timing of failure within the timeframe available for this first analysis, but there was a tendency for breeding failure of white-tailed eagles on the Western Isles to occur at an earlier stage (mostly before hatching) than in other regions, where most failures in 2022 were at chick stage (pers. obs.). This suggests that failures in the Western Isles were related to impacts on breeding adults. These could have been caused by earlier HPAI infections than those affecting chicks in other regions (possibly implying different sources of infection in these different areas, such as flocks of migrating geese in early spring, or breeding seabirds in late spring and early summer). Alternatively, failures due to poor adult condition might have been caused by other factors that were unrelated to HPAI, such as the high levels of rainfall in February which could have impacted on the condition of pre-laying, laying or incubating adults. Where failures are caused by HPAI, finding out more about the stage of failure could inform actions aimed at mitigating the risks or consequences of infection.

Immunological testing to confirm infections in live raptors could provide data that usefully complements SRMS information on breeding impacts. However, the scale at which it is feasible to carry out sampling on a professional basis is likely to be limited, and potential impacts on survival and breeding outcomes arising from extra visits to carry out serological sampling would have to be considered. From the perspective of minimising economic costs and negative impacts on welfare and conservation, enabling volunteer raptor surveyors to carry out sampling during the course of routine survey visits could be helpful. Many of these volunteers are highly skilled fieldworkers, but training and licensing large numbers of them to take blood, tissue or cloacal swab samples from live adults and chicks under Home Office licences would be challenging. Provided that AI can be detected from less invasive sampling techniques like buccal swabbing (USDA, 2023) or from faeces, pellets or prey remains collected from nest sites, these types of sampling could allow more volunteers to contribute useful information. The level of inference that will be possible from immunological testing will be considerably strengthened if it can be directly related to the occupancy and breeding productivity of the individual pairs from which serological sampling takes place.

It would be helpful to continue testing dead raptors whose cause of death might be HPAI-related in as wide a range of situations as possible – not just adults and chicks found at active nest sites, but also dead raptors found in other locations, and outside of the breeding season. As post-mortem examinations of dead raptors are often carried out to aid in detection and prevention of wildlife crime, raptors carcasses are commonly reported to authorities. Because a positive HPAI test normally precludes further post-mortem tests such as X-rays or toxicological screening, clear guidelines are required for determining when HPAI tests should be prioritised over other types of examination.

For reporting of HPAI tests on raptors and other wild bird species, it would also be useful to ensure that, as well as the results of positive tests that have been reported during the past year, information on negative test results and on the locations, age and sex of tested birds, as well as (where applicable) the variant of HPAI detected, is recorded and made as widely available as possible. It would also be helpful to extend the range of avian prey species for which carcasses are routinely tested (e.g. including feral pigeons and passerines), as this could indicate where raptors are at higher risk of infection from live prey or avian carrion.

The strength and utility of evidence communicated by SRMS data contributors would benefit from improvements to the SRMS Online data portal to facilitate systematic recording of information and evidence relating to causes of failure. This would strengthen our ability to identify links between observations of dead, dying and sick birds, as well as spatial and temporal trends in nest abandonment and breeding failures due to HPAI. The improvements to SRMS Online needed to provide users with this functionality have been scoped and costed, but delivering these will require sufficient resource for IT development of the existing system.

5. References

Animal and Plant Health Agency. 2022. Highly pathogenic avian influenza (HPAI) in the UK and Europe. Updated Outbreak Assessment #32. Ref: VITT/1200 HPAI in the UK and Europe.

Animal and Plant Health Agency. 2023. Bird flu (avian influenza): cases in wild birds. Confirmed cases of wild birds with highly pathogenic avian influenza (bird flu) in Great Britain.

BTO. 2023. BirdFacts: profiles of birds occurring in the United Kingdom. BTO, Thetford.

Challis, A., Wilson, M.W., Eaton, M.A., Stevenson, A., Stirling-Aird, P., Thornton, M. & Wilkinson, N.I. 2022. Scottish Raptor Monitoring Scheme Report 2020. BTO Scotland, Stirling.

Hardey, J., Crick, H., Wernham, C., Riley, H., Etheridge, B. & Thompson, D. 2013. Raptors: a field guide to survey and monitoring (3rd Edition). The Stationery Office, Edinburgh.

Harris, S.J., Massimino, D., Balmer, D.E., Kelly, L., Noble, D.G., Pearce-Higgins, J.W., Woodcock, P., Wotton, S. and Gillings, S. 2022. The Breeding Bird Survey 2021. BTO Research Report 745. British Trust for Ornithology, Thetford.

Hayhow, D.B., Benn, S., Stevenson, A., Stirling-Aird, P.K. & Eaton, M.A. 2017. Status of Golden Eagle Aquila chrysaetos in Britain in 2015. Bird Study 64: 281-294.

MetOffice. 2023. Climate and climate change: UK and regional series.

Newton, I. 1979. Population Ecology of Raptors. T & AD Poyser, Berkhamsted.

Pearce-Higgins, J.W., Humphreys, E.M., Burton, N.H.K., Atkinson, P.W., Pollock, C., Clewley, G.D., Johnston, D.T., O’Hanlon, N.J., Balmer, D.E., Frost, T.M., Harris, S.J. & Baker, H. 2023. Highly pathogenic avian influenza in wild birds in the United Kingdom in 2022: impacts, planning for future outbreaks, and conservation and research priorities. Report on virtual workshops held in November 2022. BTO Research Report 752, BTO, Thetford, UK.

Ratcliffe, D.A. 1993. The Peregrine Falcon. T & AD Poyser, London.

Sansom, A., Etheridge, B., Smart, J. & Roos, S. 2016. Population modelling of North Scotland red kites in relation to the cumulative impacts of wildlife crime and wind farm mortality. Scottish Natural Heritage Commissioned Report No. 904.

Tanno, S. 2022. UK weather: Spring cold snap brings flurries of snow and plummeting temperatures. The Telegraph, 1^st April 2022.

USDA. 2023. Avian sample collection for Influenza A and Newcastle Disease. USDA Animal and Plant Health Inspection Service.

Veen, J., Brouwer, J., Atkinson, P., Bilgin, C., Blew, J., Eksiovglu, S., Hoffmann, M., Nardelli, R., Spina, F., Tendi, C., & Delany, S. 2007. Ornithological data relevant to the spread of Avian Influenza in Europe (phase 2): further identification and first field assessment of Higher Risk Species. Wetlands International, Wageningen, The Netherlands.

Wilson, M.W., Challis, A. & Wernham, C.V. 2022. Scottish Raptor Monitoring Scheme Trends for 2009-2018 Methods and Analysis of Gaps (A report to the SRMG). BTO Scotland, Stirling.

Wilson, M.W., Cook, A.S.C.P., Horswill, C., Robinson, R.A. & Wernham, C.V. 2019. Population modelling for the Scottish Northern raven population. Scottish Natural Heritage Research Report No. 1012.

Woodward, I.D., Massimino, D., Hammond,M.J., Barber, L., Barimore, C., Harris, S.J., Leech, D.I., Noble, D.G., Walker, R.H., Baillie, S.R. & Robinson, R.A. 2021. Trends in numbers, breeding success and survival for UK breeding birds. BTO, Thetford, UK.