EDINBURGH NAPIER UNIVERSITY

Similar documents
Woodcock: Your Essential Brief

The grey partridges of Nine Wells: A five-year study of a square kilometre of arable land south of Addenbrooke s Hospital in Cambridge

DO DIFFERENT CLUTCH SIZES OF THE TREE SWALLOW (Tachycineta bicolor)

What is the date at which most chicks would have been expected to fledge?

Breeding Activity Peak Period Range Duration (days) Laying May May 2 to 26. Incubation Early May to mid June Early May to mid June 30 to 34

The grey partridges of Nine Wells. A study of one square kilometre of arable land south of Addenbrooke s Hospital in Cambridge

Key concepts of Article 7(4): Version 2008

Key concepts of Article 7(4): Version 2008

Breeding Activity Peak Period Range Duration (days) Egg laying Late May to early June Mid-May to mid-july 3 to 10

Survivorship. Demography and Populations. Avian life history patterns. Extremes of avian life history patterns

Adjustments In Parental Care By The European Starling (Sturnus Vulgaris): The Effect Of Female Condition

Naturalised Goose 2000

The distribution of Hen Harriers in Ireland in relation to land use cover, particularly forest cover

For further information on the biology and ecology of this species, Clarke (1995) provides a comprehensive account.

The hen harrier in England

Scottish Natural Heritage Diversionary feeding of hen harriers on grouse moors. a practical guide

Does supplementary feeding reduce predation of red grouse by hen harriers?

Breeding White Storks( Ciconia ciconia at Chessington World of Adventures Paul Wexler

Breeding Activity Peak Period Range Duration (days) Site occupation and territorial display Early April Mid-March to early May

BROOD REDUCTION IN THE CURVE-BILLED THRASHER By ROBERTE.RICKLEFS

Intraspecific relationships extra questions and answers (Extension material for Level 3 Biology Study Guide, ISBN , page 153)

Ames, IA Ames, IA (515)

BLACK OYSTERCATCHER NEST MONITORING PROTOCOL

Breeding success of Greylag Geese on the Outer Hebrides, September 2016

Egyptian vulture (Neophron percnopterus) research & monitoring Breeding Season Report- Beypazarı, Turkey

Long-eared Owl. For further information on the biology and ecology of this species, Scott (1997) provides a comprehensive account.

For further information on the biology and ecology of this species, Clarke (1996) provides a comprehensive account.

Rabbits and hares (Lagomorpha)

Demography and breeding success of Falklands skua at Sea Lion Island, Falkland Islands

Ciccaba virgata (Mottled Owl)

For further information on the biology and ecology of this species, Chapman (1999) provides a comprehensive account.

NORFOLK BIODIVERSITY ACTION PLAN Ref 1/S8 Tranche 1 Species Action Plan 8 GREY PARTRIDGE

Gambel s Quail Callipepla gambelii

Mate protection in pre-nesting Canada Geese Branta canadensis

Great Horned Owl (Bubo virginianus) Productivity and Home Range Characteristics in a Shortgrass Prairie. Rosemary A. Frank and R.

Tree Swallows (Tachycineta bicolor) are breeding earlier at Creamer s Field Migratory Waterfowl Refuge, Fairbanks, AK

Below, we present the methods used to address these objectives, our preliminary results and next steps in this multi-year project.

Vigilance Behaviour in Barnacle Geese

Executive Summary. December Page 2

Sheikh Muhammad Abdur Rashid Population ecology and management of Water Monitors, Varanus salvator (Laurenti 1768) at Sungei Buloh Wetland Reserve,

AN APPLIED CASE STUDY of the complexity of ecological systems and process: Why has Lyme disease become an epidemic in the northeastern U.S.

State birds. A comparison of the Northern Mockingbird and the Western Meadowlark. By Shaden Jensen

12 The Pest Status and Biology of the Red-billed Quelea in the Bergville-Winterton Area of South Africa

Key concepts of Article 7(4): Version 2008

PROBABLE NON-BREEDERS AMONG FEMALE BLUE GROUSE

Contrasting Response to Predator and Brood Parasite Signals in the Song Sparrow (melospiza melodia)

Summary of 2016 Field Season

769 q 2005 The Royal Society

Breeding Activity Peak Period Range Duration (days) Egg laying Mid-March to early April Early March to mid-april 3 to 6 (for clutch of 2)

DO BROWN-HEADED COWBIRDS LAY THEIR EGGS AT RANDOM IN THE NESTS OF RED-WINGED BLACKBIRDS?

Breeding Activity Peak Period Range Duration (days) Egg laying Early April Mid-March to early May 3 to 12

VIRIDOR WASTE MANAGEMENT LIMITED. Parkwood Springs Landfill, Sheffield. Reptile Survey Report

Between 1850 and 1900, human population increased, and 99% of the forest on Puerto Rico was cleared.

By Hans Frey ¹ ² & Alex Llopis ²

Hole-nesting birds. In natural conditions great and blue tits breed in holes that are made by e.g. woodpeckers

University of Canberra. This thesis is available in print format from the University of Canberra Library.

Record of Predation by Sugar Glider on Breeding Eastern Rosellas 33Km NE of Melbourne in November 2016

GUIDELINES FOR APPROPRIATE USES OF RED LIST DATA

Stray Dog Survey A report prepared for: Dogs Trust. GfK NOP. Provided by: GfK NOP Social Research. Your contact:

Development of the New Zealand strategy for local eradication of tuberculosis from wildlife and livestock

Internship Report: Raptor Conservation in Bulgaria

Hares: Ecology and Survey

Removal of Alaskan Bald Eagles for Translocation to Other States Michael J. Jacobson U.S Fish and Wildlife Service, Juneau, AK

Breeding Activity Peak Period Range Duration (days) Egg laying Late April to mid-may Mid-April to late June 5 to 12

Growth and Development. Embryonic development 2/22/2018. Timing of hatching. Hatching. Young birds and their parents

Western Snowy Plover Recovery and Habitat Restoration at Eden Landing Ecological Reserve

Population dynamics of small game. Pekka Helle Natural Resources Institute Finland Luke Oulu

Garden Birds. Blackbird Latin Name: Turdus merula

Reduced availability of refuse and breeding output in a herring gull (Larus argentatus) colony

Short Report Key-site monitoring on Hornøya in Rob Barrett & Kjell Einar Erikstad

Ecology and Management of Ruffed Grouse and American Woodcock

Piping Plover. Below: Note the color of the sand and the plover s back.

Effects of Parasitism by Brown-headed Cowbirds May Persist into Post-fledging

Barn Swallow Nest Monitoring Methods

This list of butterflies and moths is not. Acknowledgements. Further reading

KS3 Adaptation. KS3 Adaptation. Adaptation dominoes Trail

Mr T.B Brown. Land off Turweston Road, Northamptonshire REPTILE SURVEY REPORT

SEASONAL PATTERNS OF NESTING IN THE RED-WINGED BLACKBIRD MORTALITY

REGIONAL VARIATION IN COWBIRD PARASITISM OF WOOD THRUSHES

Raptor Ecology in the Thunder Basin of Northeast Wyoming

INFLUENCE OF FEED QUALITY ON THE EXPRESSION OF POST WEANING GROWTH ASBV s IN WHITE SUFFOLK LAMBS

Bird Species Fact Sheets

Quantifying density dependence in a bird population using human disturbance

The Long-term Effect of Precipitation on the Breeding Success of Golden Eagles Aquila chrysaetos homeyeri in the Judean and Negev Deserts, Israel

Co-operative breeding by Long-tailed Tits

Short Report Key-site monitoring on Hornøya in Rob Barrett & Kjell Einar Erikstad

Factors Influencing Local Recruitment in Tree Swallows, Tachycineta bicolor

29. COULD THE HEN HARRIER (CIRCUS CYANEUS) DECLINE ON ORKNEY BE DUE TO A SHORTAGE

ASPECTS OF THE BREEDING BIOLOGY AND PRODUCTIVITY OF BACHMAN S SPARROW IN CENTRAL ARKANSAS

OBSERVATIONS ON A PAIR OF NIGHTJARS AT THE NEST

Appendix 6.4. Reptile Survey

PREDATION ON RED-WINGED BLACKBIRD EGGS AND NESTLINGS

SHEEP SIRE REFERENCING SCHEMES - NEW OPPORTUNITIES FOR PEDIGREE BREEDERS AND LAMB PRODUCERS a. G. Simm and N.R. Wray

Ernst Rupp and Esteban Garrido Grupo Jaragua El Vergel #33, Santo Domingo Dominican Republic

Summary of 2017 Field Season

HOW MANY BASKETS? CLUTCH SIZES THAT MAXIMIZE ANNUAL FECUNDITY OF MULTIPLE-BROODED BIRDS

Male parental care and monogamy in snow buntings

THE BEGGING BEHAVIOR OF NESTLING EASTERN SCREECH-OWLS

The House Mouse (Mus musculus)

Texas Quail Index. Result Demonstration Report 2016

Transcription:

EDINBURGH NAPIER UNIVERSITY PROJECT REPORT: 2011 MSC CONSERVATION AND MANAGEMENT OF PROTECTED AREAS SCHOOL OF LIFE, SPORT AND SOCIAL SCIENCES David Anderson 10001884 A COMPARATIVE STUDY OF RING OUZEL (TURDUS TORQUATUS) FOOD PROVISIONING TO NESTLINGS IN A DECLINING AND AN INCREASING BREEDING POPULATION WITHIN THE CAIRNGORMS NATIONAL PARK, SCOTLAND

A COMPARATIVE STUDY OF RING OUZEL (TURDUS TORQUATUS) FOOD PROVISIONING TO NESTLINGS IN A DECLINING AND AN INCREASING BREEDING POPULATION WITHIN THE CAIRNGORMS NATIONAL PARK, SCOTLAND David Anderson Research Dissertation MSc Conservation and Management of Protected Areas School of Life, Sport and Social Sciences Edinburgh Napier University 1 September 2011 2

Contents Page 1 Acknowledgements 4 2 Abstract 5 3 Introduction 6 4 Methodology 11 -Study Area 11 -Determination of Nesting Site Occupancy 11 -Food Provisioning Watches 12 -Analyses 13 5 Results 16 -Nestling Provisioning Rates 16 -Nestling Diet 16 -Prey Load Size 19 6 Discussion 21 -Future Research and Conservation Priorities 25 7 Conclusions 27 8 Reference List 28 9 Appendix I: Glen Clunie Provisioning Watch Data 32 10 Appendix II: Glen Effock Provisioning Watch Data 33 11 Appendix III: Additional Research Photos 34 12 Appendix IV: Comparative Analysis of Glen Clunie Fed Nests 37 3

Acknowledgements Thank you to all the RSPB staff that assisted me in the completion of this project, in particular Innes Sim who has taught me a great deal over the past few months. Thank you also to Mike Nicholl of the Tay Ringing Group for sharing his knowledge of the Glen Effock study area. Thank you to the staff of the Invercauld and Invermark estates for allowing me access to the study areas. 4

Abstract The food provisioning of adult Ring Ouzels to nestlings was studied over a single breeding season in Glen Clunie and Glen Effock, in the Cairngorms National Park, Scotland. The aim of the study was to investigate whether food availability during the nesting period was influencing population trends through a comparison of a declining (Glen Clunie) and increasing (Glen Effock) breeding population. There was no significant difference in the parental provisioning rates between Glen Clunie and Glen Effock. The nestling diet was dominated by earthworms and larvae, with the latter forming a significantly higher proportion of the nestling diet in Glen Effock. Larvae was also found to increase as the nestlings got older and as the season progressed. The size of the prey load delivered to nestlings was found to be significantly higher in Glen Effock, while also increasing over the course of the season. The lack of difference in parental provisioning rates suggests that adult Ring Ouzels are able to provision nestlings adequately under a range of conditions. The difference in prey load size between the two sites may have an effect on early brood partial mortality thereby impacting upon population trends. There is also evidence that larvae is more abundant in Glen Effock and therefore may be buffering nestlings in this area from the effects of declining earthworm availability. Overall however, the differences in prey load and prey type are not enough, in the context of no difference in provisioning rates, to provide clear evidence that population trends in Glen Clunie and Glen Effock are significantly affected by food availability during the nesting period. 5

Introduction The Ring Ouzel Turdus torquatus is a summer visitor to the United Kingdom, and is primarily a bird of moorland and montane areas (Wotton, Langston & Gregory, 2002). It nests mainly on steep sided valleys, on crags and in gullies (Sim et al, 2010) with nesting pairs being found from near sea level in the far north of Scotland, up to 1,200 metres in the Cairngorms (Gibbons, Reid & Chapman, 1993; Wotton, Langston & Gregory, 2002). Within these areas it has been found that breeding densities correlate positively with areas of undifferentiated heather containing patches of short grass (Buchanan et al, 2003) (Figure 1). These heather-grass mosaics provide the breeding pairs with prime foraging sites to find the invertebrates upon which the young feed during the nesting period, while also providing sufficient cover from predators. Figure 1. A typical Ring Ouzel nesting habitat located in Glen Effock, June 2011. Note the steep banks with plenty of heather cover for nests and fledglings. There are also large areas of short grass that provide prime foraging sites in which provisioning adults can find the invertebrates upon which the nestlings feed. Once widespread during the 19 th century, with breeding records from the south of England to as far north as Orkney, the breeding population of Ring Ouzels has since gone into a long 6

term decline (Sim et al, 2010). The decline was first noted at the beginning of the 20 th century (Gibbons, Avery & Brown, 1996) and subsequent breeding bird atlases have shown a continuing population decrease, with a 27% reduction in the breeding range of the species between the 1968-1972 and 1988-1991 national atlases (Sharrock, 1976; Gibbons, Avery & Brown, 1996). Recent decades have shown no sign of this decline halting with the first national survey of the species in 1999 putting the population at 6,157-7,549 pairs (Wotton, Langston & Gregory, 2002). While this is within the range of the 1988-1991 atlas, it still showed a 39-43% decline in the tetrad occupancy. Further to this, re-analysis of the data has suggested a population decline of 58% between 1991 and 1999 (Buchanan et al, 2003), marking out the Ring Ouzel as a species of serious conservation concern. Due to this concern the Ring Ouzel has recently been red-listed and made a priority species in the UK Biodiversity Action Plan (Sim et al, 2010). As such it is vital that the reasons for this decline are clearly understood in order that appropriate conservation action can be taken. This however has proved difficult, with many possible causes being put forward to explain the decline of the species. It is only recently that sufficient evidence has begun to emerge to indicate which of these may be important and in what stage of the life cycle they are operating. Studies by Sim et al (2007) and Buchanan et al (2003) have indicated that Ring Ouzels breeding distribution has contracted to areas at higher altitudes with greater heather cover (Figure 2), while also moving away from conifer plantations. It is unclear however whether the relationships outlined above are casual or are simply a reflection of the species contracting to preferred breeding sites as the population declines. As Sim et al (2010) argue other studies have shown declines where there has been no significant afforestation or change in heather cover; therefore it seems that, though such land use changes may be important locally, it is unlikely that they are a major driver for the overall decline within the UK. Climate correlates have also been examined by Beale et al (2006) to investigate whether the decline of the Ring Ouzels in the UK may be linked to climate change. It was found that population declines occurred following British summers that were warm and wet, and when spring rainfall was high in the Ring Ouzels wintering grounds 24 months previously. This led 7

Figure 2. A typical Ring Ouzel nest with a clutch of four eggs in Glen Effock, June 2011. The nest is located in deep heather to provide the eggs and nestlings with sufficient cover to prevent predation. Significant change in heather cover could therefore affect the ability of Ring Ouzels to nest successfully. to the hypothesis that a combination of these factors was lowering food availability in both breeding and wintering grounds, thereby impacting upon population trends. However, as Sim et al (2010) state, it would require a longer run of data to assess whether climate variables are impacting upon local Ring Ouzel survival and productivity. It therefore remains unclear as to the impact of climate change on the breeding population of Ring Ouzels in the UK. A long term study of a declining breeding population of Ring Ouzels in Glen Clunie in the Cairngorms National Park has begun to shed some more light on the causes of the decline. The study of this population has shown no decline in adult survival over twelve years, however there has been a marked reduction in the return rates of first year birds (Sim et al, 2010). As the Ring Ouzel is a migrant, these low survival rates could be due to factors affecting breeding grounds, overwintering areas, or migration routes. Fennoscandia populations however, which share wintering grounds with the UK breeding population, have not shown the same declines as those in the UK (Bird Life International, 2004) suggesting that it is factors out-with wintering grounds that are causing this poor survival rate. 8

The breeding data from Fennoscandia is sparse and often of poor quality, as such any conclusions drawn from this data has to be treated with caution. A study by Burfield and Brooke (2005) however used migration routes of Fennoscandia Ring Ouzels as an alternative way of comparing this population with that in the UK. The migration route of Fennoscandia birds includes the east coast of the UK whereas British breeders tend to migrate over the west coast. The study showed that counts at west coast observatories declined but those at east coast observatories did not, indicating that the Fennoscandia population was not experiencing the same decline as seen in the UK. Further evidence of the importance of breeding grounds comes from a recent study by Sim et al (2010) that examined recent population trends across fourteen different breeding areas throughout the UK. In keeping with an overall declining population trend, eleven areas showed significant declines over 50%, and in two areas the population had declined to extinction. The one exception to this was the study area in Glen Effock where, rather than declining, the population has increased by 88% over six years. When compared to the declining population in Glen Clunie, which is located only 30 km to the north-west, the data showed that chick return rates were significantly higher in Glen Effock (Sim et al, unpubl. data). This supports the assertion that it is factors affecting breeding grounds that is driving the population trend. It is of vital importance that these factors are researched as they may be key to arresting the decline of the population within the UK. One factor that could affect Ring Ouzels during this period is that of decreased reproductive success due to partial brood loss through starvation. This could occur if food availability on breeding grounds is such that adults are not able to adequately provision nestlings. Further to this, low food availability during the nesting period could cause the nestlings to fledge the nest in poor body condition, and thereby increase their vulnerability to predation. It is hypothesised that low food availability on breeding grounds during the nesting period is having an adverse impact upon reproductive success thereby contributing to the population decline. This study aimed to test this hypothesis by comparing food provisioning to nestlings in a declining (Glen Clunie) and increasing (Glen Effock) Ring Ouzel population. To achieve this aim food provisioning rates to nestlings in both study areas were recorded and compared. The type of prey and size of the prey load were also recorded to investigate whether there was any difference in the content of the feeds. Together these variables provide an indication 9

of the availability of food in both Glen Clunie and Glen Effock and whether this factor is driving the differing population trends in the two study areas. 10

Methodology Study Area The study was carried out at Glen Clunie (Figure 3; 56⁰56 N, 3⁰25 W) and Glen Effock (Figure 4; 56⁰53 N, 2⁰54 W) in the Cairngorms National Park during the spring and summer of 2011 (April-July). Both areas represent typical Ring Ouzel habitat in the UK, comprising as they do of heather moorland containing patches of short grass. They have also been used as Ring Ouzel study areas (Sim et al, 2010) over the past decade and therefore recent population trends in both areas are well known. The management of each study area is also similar, with predators such as Fox (Vulpes vulpes), Stoat (Mustela erminea) and Crows (Corvus spp.) all controlled in order to manage the sites for Red Grouse (Lagopus lagopus scoticus) shooting. Further to this, rotational burning of heather is also employed in both sites to maintain suitable habitat for all stages of the Red Grouse life cycle. Crown copyright/database right 2011. An Ordnance Survey/EDINA supplied service Figure 3. Map of Glen Clunie study area (scale 1:100000) 11

Crown copyright/database right 2011. An Ordnance Survey/EDINA supplied service Figure 4. Map of Glen Effock study area (scale 1:40000) Determination of Nesting Site Occupancy Data on the location of previously occupied nesting territories was known from earlier research conducted in both study areas, however surveying of each site was required in order to locate the exact nesting sites. These surveys involved walking through suitable habitat in each study area watching for foraging birds and listening for singing males. In Glen Clunie; tape playback of Ring Ouzel song was used to elicit a response from males in the area and enhance the chances of locating a nest (Gilbert, Gibbons & Evans, 1998). When a bird was located it was followed using binoculars so that the exact location of the nest could be identified, this allowed for a closer inspection to be made on foot; enabling the exact stage of the breeding cycle to be identified. The GPS co-ordinates of each nest were then noted and subsequent visits to the nests enabled exact laying and hatching dates to be recorded. All nests found were included in the research if possible, however in Glen Clunie simultaneous research was being undertaken to artificially feed several nests thereby automatically ruling these nests out of this study. Other nests had to be discounted due to access difficulties (such as the nest being located on an inaccessible crag) or due to the nest being too close to a heavily used footpath, or road, which could cause disturbance. These problems represented more of an issue in Glen Clunie as the study area is dissected by a busy 12

road and many footpaths radiate out in to the surrounding area. Glen Effock on the other hand is rarely used by ramblers and disturbance of the nests was therefore not an issue. Food Provisioning Watches Food provisioning watches were conducted at nests in both Glen Clunie (n = 18) and Glen Effock (n = 13) after the chicks reached seven days old. These watches were conducted using a portable hide (Appendix 3), binoculars (x10) and a spotting scope (x20-x60). Before each watch; the nest was monitored for thirty minutes to indicate from which direction the adults usually entered the nest, thereby assisting in the selection of the location for the hide. The nest was then approached and checked to confirm the size of the brood, after which the hide was set up, ideally within twenty metres of the nest to allow the type of prey being brought in to be identified. After the hide was set up, depending on the character of the adult pair, it was sometimes required to withdraw from the hide and try and enter it when both adults were away from the nest. On most occasions however this was not required and the hide could be entered directly after it was set up. After entering the hide, the adult pair was monitored until it was certain that their behaviour was not being affected by the presence of the hide. To ensure this, the watch was not started until both the adult male and female had entered the nest with food. The watch then commenced and lasted for two hours with the number of feeds, prey type and the size of the prey load for each feed all recorded during that time. Due to the character of some birds and the location of some nests it was not always possible to record the prey type and size for every feed. When possible however, the type of prey was categorised as earthworm, unidentified larvae or caterpillar, with other prey types identified when possible. The total size of each prey load brought to the nest was recorded in relation to the length of the bill (Chamberlain, Hatchwell & Perrins, 1999). Further to this several other variables were also recorded, including size of the brood, age of the nestlings and time of day, to ensure that interactions between different variables could be examined. At some nest sites the breeding pair did not settle after the hide was set up, on these occasions the watch was abandoned and attempted again on subsequent days. If the birds again failed to settle the hide was moved well back from the nest and a watch conducted that recorded the feeding rate but excluded the prey type and size, due to the hide being too far away for reliable data. 13

Analyses All analysis of data recorded during the food provisioning watches was undertaken using SAS statistical software. Each data set was analysed using the Anderson-Darling test to ensure that it conformed to a normal distribution that would thereby allow the use of parametric tests. If the data did not conform to a normal distribution, standard data transformations were employed to ensure that the analyses could be conducted without the need for non-parametric tests. The feeding rate recorded during the nest provisioning watches was analysed to test for any significant difference (P < 0.05) in the number of feeds per watch between Glen Clunie and Glen Effock. The dependent variable was the number of feeds observed during the two hour watch (which exhibited a normal distribution) with each watch included as a single data point. These data points however were not all independent as multiple watches were conducted on some nests and many of the same territories were used again for watches on second broods. A General Linear Mixed Model (GLIMMIX) procedure was therefore used to enable normally distributed random effects to be incorporated. This involved specifying territory number as a random effect, thus enabling the analysis to take account of any nonindependent variables and reduce the degrees of freedom by the appropriate amount. The explanatory variables initially used in the model were site (Glen Clunie; Glen Effock), brood (early; late), time of watch (morning; early afternoon; late afternoon), nestling age, size of brood and day of the season (all co-variates). Interaction terms were also included in the model to ensure any correlation between explanatory variables was included in the analysis. The model was then selected through the backwards deletion of non-significant (P > 0.05) terms to indicate whether there was any significant interaction between the study area and the feeding rate, and whether any trends over the course of the season could be viewed as significant. The prey type data was analysed to determine whether there was any significant difference (P < 0.05) in the types of prey delivered to nestlings, over the course of the breeding season, between the two study areas. During the food provisioning watches, the occurrence or nonoccurrence of each prey type (earthworms, larvae or caterpillar) was recorded for every food delivery to the nest. This data was then transformed to show the proportion of food deliveries within each watch period where a certain prey type had been recorded. This transformation 14

enabled proportional data to be produced that allowed for a direct comparison between the two study areas. The proportional data obtained through the food provisioning watches was then transformed using the arcsine square root transformation to ensure a normal distribution. The data was then put through a mixed model, which, as with the feeding rate data above, enabled the analysis to account for the possible non-independence of multiple watches conducted at the same territory. The dependent variable used in the model was the proportional prey type data, with the same explanatory variables utilised for the feeding rate analysis, used again in this model. Finally, the size of the prey loads delivered to the nestlings were analysed to determine whether there was any significant difference between the study areas and whether any trends over the course of the season could be viewed as significant. The total size of each prey load delivered to the nest was recorded in relation to the size of the bill (eg. 1 x the length, 2 x the length, etc.) with the mean size of each prey load for each watch then calculated for use in the analysis. This data was used as the indicator variable in a mixed model as with the other analyses above. Again, this enabled the non-independence of nest sites to be accounted for and allowed the analysis to be conducted using the same six explanatory variables as outlined above. 15

No. Feeds per Watch Results Nest Provisioning Rates No significant difference was detected in the mean food provisioning rates to nestlings between Glen Clunie and Glen Effock (Figure 5; Clunie = 9.389 ± 0.662, n =18; Effock = 10.231 ± 0.818, n = 13; df = 26, F = 0.84, P = 0.367). No significant correlation was identified with any of the other explanatory variables. 17.5 15.0 12.5 10.0 7.5 5.0 Clunie Study Area Effock Figure 5. Feeding rate of Ring Ouzels nesting in Glen Clunie and Glen Effock in the Cairngorms National Park, Scotland, 2011. Nestling Diet The overall contribution of each prey type to the diet of nestlings in both study areas is set out in Table 1. Six prey types were identified; however the frequencies of dragonfly, beetle and moth prey types were at such a low level that they could not be used in the analysis of nestling diet. No significant correlation was found between proportion of earthworms in the nestling diet and any of the explanatory variables analysed. There was however a near significant correlation with study area (Clunie = 0.361 ± 0.064, n = 18; Effock = 0.294 ± 0.192, n = 13; df = 17.38, F = 4.01, P = 0.061) with Glen Clunie having a slightly higher 16

proportion of earthworms in the diet of nestlings, thus suggesting that earthworms may be marginally more common in Glen Clunie. Table 1. Number of feeds where each prey type was present in each study area Prey Type No. of Feeds: Clunie No. of Feeds: Effock Earthworm 62 41 Caterpillar 28 18 Unidentified Larvae 37 57 Dragonfly 1 0 Beetle 1 0 Moth 3 2 Caterpillars were regularly delivered to nestlings in both study areas, however no significant difference was observed in the proportion that they formed of the nestling s diet between Glen Clunie and Glen Effock (Clunie = 0.162 ± 0.040, n = 18; Effock = 0.147 ± 0.029, n = 13; df = 2.55, F = 0.01, P = 0.927). No other explanatory variables showed a significant correlation with the proportion of caterpillars, suggesting that the delivery of this prey type remained constant throughout the course of the season. Unlike earthworms and caterpillars however, the proportion of larvae in the nestling s diet did show significant correlations with day of the season (df = 21, F = 17.08, P < 0.001), age of the nestlings (df = 21, F = 7.29, P = 0.09) and study area (Clunie = 0.322 ± 0.063, n = 18; Effock = 0.446 ± 0.067, n = 13; df = 21, F = 7.29, P = 0.013). Larvae became more important in the nestling s diet as the season progressed peaking around the end of June and the beginning of July (Figure 6). There were two watches in late May that showed the proportion of larvae at the same level as these peak times, however in general most watches in May only recorded larvae in 30% of the feeds. Though the proportion of feeds containing larvae across the different ages of nestlings was variable, it never fell below 30% for nestlings over ten days old. Further to this, chicks under ten days old were often only fed larvae in only around 10% to 20% of feeds, therefore indicating that larvae becomes more important as the nestlings grow reaching a peak just prior to fledging at 13 days (Figure 6). 17

% of Feeds Containing Larvae % of Feeds Containing Larvae per Watch % of Feeds Containing Larvae per Watch 90 90 80 80 70 70 60 60 50 50 40 40 30 30 20 20 10 10 0 10 30 50 70 0 6 8 10 12 Day of Season Age of Nestlings Figure 6. Proportion of larvae in nestling s diet expressed as the percent of feeds per watch that contained larvae plotted against day of the season and age of the nestlings with straight trend line. Both relationships showed a significant positive trend. Larvae also proved to be far bigger part of the nestling diet in Glen Effock than they were in Glen Clunie. As is indicated in Figure 7, the majority of the watches in Glen Clunie contained larvae in less than 30% of the observed feeds which contrasted with Glen Effock where, though the proportion of larvae was more variable, the majority was over 50%. Further to this there was a marked difference in the mean proportion of feeds where larvae was observed, with Glen Clunie (32.17%) being over 10% lower than Glen Effock (44.62%). It therefore appears that larvae are far more common in Glen Effock than they are in Glen Clunie. 80 70 60 50 40 30 20 10 0 Clunie Study Area Effock Figure 7. Boxplot showing the percentage of feeds where larvae was present per watch in Glen Clunie and Glen Effock in the Cairngorms National Park, Scotland, 2011. 18

Mean Prey Load Size per Watch (x length of beak) Prey Load Size The analysis carried out on prey load size relates to the overall size of the prey load and not the individual prey items, therefore no conclusions can be drawn about relative sizes of earthworms, larvae etc. The size of the prey load delivered to the nestlings showed significant correlations with day of the season (df = 27, F = 8.68, P = 0.007) and with study area (Clunie = 3.380 ± 0.263, n = 18; Effock = 4.187 ± 0.160, n = 13; df = 27, F = 7.08, P = 0.013), however none of the other explanatory variables analysed showed any significant correlation. The prey load size grew steadily as the season progressed peaking around the end of June and beginning of July indicating that later broods received a larger prey load per feed than those earlier season (Figure 8). 6.00 5.00 4.00 3.00 2.00 1.00 0.00 0 10 20 30 40 50 60 70 80 Day of the Season Figure 8. Seasonal variation in mean prey load size per watch with straight trend line. Prey load size measured in relation to the size of the beak The prey load size also proved to be consistently bigger in Glen Effock than it was in Glen Clunie. In Glen Clunie the mean prey load size was 3.38 times the size of the bill, compared to 4.19 times the size of the bill in Glen Effock. The variability of the size of the prey load was also far greater in Glen Clunie. This is displayed clearly in Figure 9 which shows that the mean prey load size for each watch in Glen Effock never fell below 3 times the size of the bill where-as in Glen Clunie, several watches had a mean prey load size of 3 times the size of the bill or less. 19

Mean Prey Load Size per Watch (x size of beak) 6 5 4 3 2 1 Clunie Study Area Effock Figure 9. Mean prey load size per watch in Glen Clunie and Glen Effock in the Cairngorms National Park, Scotland, 2011. 20

Discussion The results of this study indicate that the Ring Ouzel population in Glen Clunie does not provision nestlings with food at a significantly lower rate than the Glen Effock population. These results mirror those from a study on another thrush species, the blackbird Turdus merula, by Chamberlain, Hatchwell and Perrins (1999) which found that parental provisioning rates did not differ between farmland and woodland habitats, even though the farmland population was exhibiting demographic characteristics of a declining population. It therefore appears unlikely that the rate of food provisioning to nestlings has any effect on the population trend in Glen Clunie. The results outlined above are contrary to the expected outcomes of this study as it was hypothesised that the declining population in Glen Clunie would exhibit evidence of lower food availability through decreased parental provisioning rates. One possible reason as to why no difference was found between the sites is that the adult birds may compensate for lower food availability with a higher feeding effort (Buehler et al, 2002). A study by Prigmore (2003) found that Ring Ouzel foraging sites tend to be within 450m of the nest, however when necessary, they can travel considerable distances (up to 1.95km) to find food with which to provision their young. Further to this, Peach et al (2004) found that provisioning Song Thrush Turdus philomelos adults will travel a greater distance from nests during periods of low food availability. Therefore, it is possible that parents in Glen Clunie may have been traveling further, and foraging for longer, than parents in Glen Effock in order to achieve the same provisioning rates. Any increased feeding effort in Glen Clunie may well have an effect on adult survival, as it would increase energy expenditure by the parents. The study of blackbirds by Chamberlain, Hatchwell and Perrins (1999) argued this point stating that this increased reproductive cost may affect subsequent survival or fecundity therefore causing the adults to bear the cost of reproduction in an area of low prey availability. Adult survival rates of Ring Ouzels in Glen Clunie are low compared to similar species in Britain and North America (Sim et al, 2010) thus suggesting that the reproductive cost of increased feeding effort may be affecting adults in this area. The impact of adult survival on the overall population trend of the Ring Ouzel however is unclear as other analyses do indicate that first year survival is the main cause of variation in 21

population growth of Ring Ouzels (Sim et al, 2011; Sim et al, 2010). As such, even if this increased reproductive cost to adults is occurring it may not be enough to explain the differing population trends of Glen Clunie and Glen Effock. None the less, adult survival cannot be ruled out as a possible driver of the decline and therefore the energetic impact of increased feeding effort in Ring Ouzels merits further study. It is possible that weather variables could have affected the parental provisioning rates that were occurring in both sites; however it was not within the scope of this study to account for these factors. The potential importance of weather variables is indicated by a study on Great Tits Parus major by Keller and van Noordwijk (1994). This study found that rainfall had an effect on the feeding frequency of parents thereby impacting upon the daily growth of the nestlings. Further to this, Stokke et al (2005) argued that bad weather during the breeding season is most likely to influence food delivery to nestlings. Though Glen Clunie and Glen Effock are only 30km apart, it is still a sufficient distance for weather variables to differ significantly and as such influence results. It has to be stated however that the study by Chamberlain, Hatchwell and Perrins (1999) on blackbirds found only a very weak effect of weather variables on parental provisioning rates thus suggesting that for thrush species, such as the Ring Ouzel, this may not be an important factor. The study was also unable to account for the influence of competition through increased population density. It is possible that the Ring Ouzel density in Glen Effock has increased competition for resources to such an extent that parental provisioning rates have become restricted thereby influencing the results of this study. This however is not likely to have been a major factor as Ring Ouzels do tend to nest in relatively low densities (< 6 pairs km²) (Burfield, 2002). As such, density dependence is more likely to be an outcome of increased predation risk (through pairs nesting in more marginal habitats) rather than increased competition (Beale et al, 2006). It therefore appears unlikely that there is any difference in the provisioning rates between Glen Clunie and Glen Effock and, though there may be an effect on adults through increased reproductive cost, it is unlikely reproductive success will be affected due to this factor. The study however also showed that the size of the prey load delivered to nestlings was significantly bigger in Glen Effock than it was in Glen Clunie. This is contrary to the results from studies on other passerine species that suggested the number of provisioning trips was a good reflection of the total mass of food delivered to nestlings (Nolan et al, 2001; MacColl & 22

Hatchwell, 2003), a relationship that is not shown in this study. This larger prey load in Glen Effock may have decreased the chance of nestling mortality due to starvation and, further to this, may have enabled nestlings to fledge the nest in better condition and as such have a better chance of survival. A further benefit of increased prey load size is that it may enable some chicks to fledge the nest early and thus escape predation events (Suedkamp-Wells et al, 2007). An occurrence of this was noted in Glen Effock where two predated nestlings were found near the nest site, the third nestling however was found 50 metres away in the heather still being fed by the parents. At only twelve days old, the nestling had obviously been able to fledge the nest early and had, as such, been able to avoid the predation event. The size of the prey load was also found to increase over the course of the season with adults able to provision more prey per feed as the year progressed. This suggests that a decrease in prey availability may have been having more of an effect on early season nestlings. This supports findings by Sim et al (2011) that showed a significant increase in partial brood mortality amongst early broods in Glen Clunie over twelve years, which led to a progressive, near significant decline in brood size at fledging. These partial brood losses were due to starvation and may reflect the smaller prey loads that parents have been able to deliver at this time of year. If the above impacts of lower prey load size over the course of the season are put in to the context of differences between the two study areas, then it may indicate reasons for the declining population in Glen Clunie. The larger prey load size overall in Glen Effock may buffer nestlings from the impacts of a smaller prey load size early in the season. However in Glen Clunie, the smaller prey load size overall may mean that any reduction early in the season will cause partial brood mortality, thereby impacting upon the overall population trend. It is not clear however whether this increase in food load is enough to sufficiently increase the growth of chicks to the extent that nestlings in Glen Effock, and those that hatch later in the season, are fledging the nest in better condition. A study by Martin et al (2011) examined the growth rate variation in 64 passerine species in both tropical and temperate sites and found that increases in food load did not fully compensate for declining number of trips among species. Furthermore, it was found that even if food loads did fully compensate for decreased provisioning rates this would not explain differing growth rates due to food 23

limitation. This indicates that even with an increased prey load size, it is unlikely that nestlings in Glen Effock would have a sufficiently higher growth rate to enable them to fledge the nest in better condition. Provisioning rates and prey load size are both important aspects of parental provisioning to nestlings, however the quality of the prey delivered in these feeds also has to be considered as this may impact upon reproductive success. Many studies cite the importance of earthworms in the diet of Ring Ouzels (Arthur & White, 2001; Rebecca, 2001) and this study generally supports this assertion with earthworms forming a large part of the nestling diet in both study areas. Larvae and caterpillars however also regularly occurred in feeds, thus suggesting that Ring Ouzels nestlings can have quite a varied diet. The contribution of larvae to the nestling diet was far greater in Glen Effock, thus suggesting that Ring Ouzel nestlings in this study area are less reliant on earthworms than they are in Glen Clunie. It has been suggested Tyler and Green (1994) that a lack of earthworms may drive adults to forage further away from the nest however this study indicates that if there is a lower abundance of earthworms, adult birds will simply switch to an alternative prey type, in this case larvae. The study was not able to show whether having this greater proportion of larvae in the diet was adversely affecting nestlings, however the fact that it was occurring in the study area with the increasing population seems to indicate that this is not an issue. Further to this, the study on blackbirds by Chamberlain, Hatchwell and Perrins (1999) supports this assertion as it was found that nestlings fed predominantly on caterpillars rather than earthworms showed only a 0.5% reduction in the nestling s survival probability to the next breeding season. The above results suggest that adults in Glen Effock may be more able to replace earthworms with other prey items if conditions dictate it necessary with little apparent cost to the nestlings. This ability may be increasingly important due to the continuing changes in climate affecting the abundance of invertebrate prey. This issue is indicated clearly in the decline of Song Thrushes Turdus philomelos in the UK where it has been shown that dry summers have serious demographic influences, as they create dry soils thereby reducing the availability of key prey items such as earthworms (Peach, Robinson & Murray, 2004). Beale et al (2006) suggested that a similar demographic influence may be occurring in Ring Ouzel populations, with higher summer temperatures on breeding grounds reducing the availability of earthworms and thereby impacting upon population trends. 24

The importance of this ability to switch to different prey types is further indicated through the increasing importance of larvae in the diet as the season progresses. As is shown above, the prey load data indicates that adults also increase the amount of prey delivered to nestlings per feed later in the season. The similar trend of increasing larvae provisioning over the season suggest that it is this prey type that is contributing most towards the increase in prey load size. This may indicate an inability of adults to increase their provisioning of earthworms thereby forcing them to bring in alternative prey types. As such the ability to switch from earthworms to alternative prey, and thereby compensate for drier soils, may prove vital if further increases in temperature and drier summers continue to exacerbate this problem. Further to this, the analysis indicates that larvae are more important in the diet of older nestlings and as such this ability to bring in more larvae just prior to fledging may be vital for chicks during the post fledging period. It is therefore possible that the greater abundance of larvae in Glen Effock has provided adults in this area with more alternative prey types thereby buffering nestlings from the effect of increasing temperatures and improving reproductive success in this area. The climate in Britain is however continuing to change with increases of 3.5⁰C possible by the end of this century (Hulme et al, 2002). As such, further work has to be done to examine the effects of soil moisture in Ring Ouzel habitats on earthworms and other invertebrate prey to gain a greater understanding of how these changes may be impacting upon population trends. Future Research and Conservation Priorities This study initially set out to examine whether differences in food provisioning to nestlings in a declining (Glen Clunie) and increasing (Glen Effock) Ring Ouzel population was impacting upon reproductive success and thereby driving the population trends. The results indicate that the food availability to nestlings in the two study areas is only subtly different, with variances in the prey load and type set against a background of no overall difference in feeding rate. This research therefore provides no firm evidence that food availability during the nesting period is impacting upon the population trends of Glen Clunie and Glen Effock. There is however some evidence that the increase in partial brood mortality amongst early broods in Glen Clunie may be caused by a lower prey load delivery in this area, possibly driven by an inability of adults to switch prey type from earthworms to larvae. It is therefore vital that the 25

abundance of different prey types in breeding grounds is analysed through sampling of foraging areas. Further to this, the food provisioning study should also be extended over several seasons to enable a greater understanding of how parental food provisioning varies under different conditions. It does however seem likely that the extrapolation of nesting data to infer population effects may not be warranted in this case and that the differences between Glen Clunie and Glen Effock may lie in other stages of the breeding season. If it is not reproductive success that is driving the decline then the logical leap is to examine the period after fledging as the effects of food availability may be more pronounced during this time. Unfortunately the post-fledging period of passerines is one of the least studied stages of avian ecology (Anders & Marshall, 2005), possibly due to fledglings secretive behaviour (something certainly true of Ring Ouzel fledglings) making repeated observations difficult (Schmidt, Rush & Ostfeld, 2008). In the case of thrush species however, this situation is slowly changing and recent years have seen several studies that have examined fledglings during this period with pressures such as predation rates (Anders et al, 1997; Schmidt, Rush & Ostfeld, 2008), changing agriculture practices (Robinson et al, 2004), ability to forage (White et al, 2005) and habitat requirements (Peach et al, 2004; Peach, Roninson & Murray, 2004) all shown to have impacted upon fledgling survival. It is therefore possible that the same pressures affecting other thrush species during the postfledging period could also be impacting upon Ring Ouzels in the UK. This has already been suggested by several studies (Beale et al, 2006; Sim et al, 2010; Sim et al, 2011) and as such a prudent next step would be to study fledglings during this time to gain a greater understanding of what pressures are affecting their survival. It is possible that the body condition of nestlings at fledging may impact upon their ability to avoid predation and as such affect post-fledging survival. Unfortunately this study was unable to compare body condition of nestlings in Glen Clunie and Glen Effock, therefore any further comparison of these populations should include this analysis. As is the case with many declining species however, it may not be possible to wait until all the questions are answered before decisive conservation action is taken. Any land management aimed at conserving Ring Ouzels should therefore ensure that needs of fledglings are accounted for as actions targeted solely at improving nest success and adult survival will not necessarily protect resources important to fledglings. 26

Conclusions Analysis of parental provisioning rates suggests that there is no significant difference between Glen Clunie and Glen Effock. This indicates that parents in both study areas are able to adequately provision their broods under a range of conditions. There was however a significant difference in the prey load size delivered to nestlings between the two study areas while both study areas also showed lower prey load sizes earlier in the season. This may explain the increase in partial brood mortality of early broods found in Glen Clunie over the past twelve years. Glen Effock may have been buffered from this effect due to the overall bigger prey loads throughout the season enabling early broods to withstand the lower prey load early in the season. This study has also found evidence that larvae are more abundant in Glen Effock and later in the season, suggesting it is this, rather than other prey types, which is driving the difference in prey load size. Other studies have indicated that earthworms are the preferred prey type of Ring Ouzels however this study indicates that an ability to switch to alternate prey types, such as larvae, may be vital in buffering broods from partial mortality. It is also possible that if earthworms continue to become less accessible due to warmer temperatures creating drier soils then larvae may be an increasingly important prey type, and areas with a greater abundance of this prey type, such as Glen Effock, may become preferred habitats for breeding Ring Ouzels. The above differences noted in prey load size and prey type, while in need of further study, are not strong enough on their own to provide clear evidence that the overall population decline in Glen Clunie is being caused by low food availability. The backdrop of a lack of difference in overall provisioning rates suggest that, while the other variables may be creating subtle differences, they may not be sufficient enough to prevent adults successfully raising their chicks to fledging. As such, this study provides no clear evidence that food availability on breeding grounds is impacting upon reproductive success and thereby driving the starkly different population trends in Glen Effock and Glen Clunie. This suggests that the factors influencing the population decline may be occurring during the post fledging period and future research should focus on studying fledglings during this time. 27

Reference List Anders, A.D., Dearborn, D.C., Faaborg, J., & Thompson III, F.R., (1997). Juvenile survival in a population of neotropical migrant birds, Conservation Biology. 11 (3), pp.698-707. Anders, A.D., & Marshall, M.R., (2005). Increasing the accuracy of productivity and survival estimates in assessing landbird population status, Conservation Biology. 19 (1), pp.66-74. Arthur, D.S.C., & White, S.A., (2001). Numbers, distribution and breeding biology of Ring Ouzel in upper Glen Esk, 1992-98, Scottish Birds. 22, pp.50-59. Beale, C.M., Burfield, I.J., Sim, I.M.W., Rebecca, G.W., Pearce-Higgins, J.W., & Grant, M.C., (2006). Climate change may account for the decline in British Ring Ouzels Turdus torquatus, Journal of Animal Ecology. 75, pp.826-835. Bird Life International, (2004). Birds in Europe: population estimates, trends and conservation status. Cambridge, Bird Life International. Buchanan, G.M., Pearce-Higgins, J.W., Wotton, S.R., Grant, M.C. & Whitfield, D.P., (2003). Correlates of the change in Ring Ouzel Turdus torquatus abundance in Scotland from 1988-91 to 1999, Bird Study. 50, pp.97-105. Buehler, D.M., Norris, D.R., Stutchbury, B.J.M & Kopysh, N.C., (2002). Food supply and parental feeding rates of Hooded Warblers in Forest Fragments, Wilson Bulletin. 114 (1), pp.122-127. Burfield, I.J., (2002). The breeding ecology and conservation of the Ring Ouzel Turdus torquatus in Britain. PhD, University of Cambridge. Burfield, I.J. & Brooke, M.deL., (2005). The decline of the Ring Ouzel Turdus torquatus in Britain: evidence from bird observatory data, Ringing and Migration. 22, pp.199-204. Chamberlain, D.E., Hatchwell, B.J. & Perrins, C.M., (1999). Importance of feeding ecology to the reproductive success of Blackbirds Turdus merula nesting in rural habitats, Ibis. 141, pp.415-427. Gibbons, D.W., Avery, M.I., & Brown, A.F., (1996). Population trends of breeding birds in the United Kingdom since 1800, British Birds. 89, pp.291-305. 28

Gibbons, D.W., Reid, J.B., & Chapman, R.A., (1993). The New Atlas of Breeding Birds in Britain and Ireland: 1988-1991. London, Poyser. Gilbert, G., Gibbons, D.W., & Evans, J., (1998). Bird Monitoring Methods: A manual of techniques for key UK species. Sandy, RSPB. Hulme, M., Turnpenny, J., & Jenkins,G., (2002). Climate change scenarios for the United Kingdom: the UKCIP02 Briefing Report. Norwich, Tyndell Centre for Climate Change Research, University of East Anglia. Keller, L.F. & Van Noordwijk, A.J., (1994). Effects of local environmental conditions on nestling growth in the Great Tit Parus major, Ardea. 82 (2), pp.349-362. MacColl, A.D.C, & Hatchwell, B.J., (2003). Heritability of parental effort in a passerine bird, Evolution. 57 (9), pp.2191-2195. Martin, T.E., Lloyd, P., Bosque, C., Barton, D.C., Biancucci, A.L., Cheng, Y.R., & Ton, R., (2011). Growth rate variation among passerine species in tropical and temperate sites: an antagonistic interaction between parental food provisioning and nest predation risk, Evolution. 65 (6), pp.1607-1622. Nolan, P.M., Stoehr, A.M., Hill, G.E., & McGraw, K.J., (2001). The number of provisioning visits by House Finches predicts the mass of food delivered, The Condor. 103, pp.851-855. Peach, W.J., Denny, M., Cotton, P.A., Hill, I.F., Gruar, D., Barritt, D., Impey, A., & Mallord, J., (2004). Habitat selection by Song Thrushes in stable and declining farmland populations, Journal of Applied Ecology. 41, pp.275-293. Peach, W.J., Robinson, R.A. & Murray, K.A., (2004). Demographic and environmental causes of the decline of rural Song Thrush Turdus philomelos in lowland Britain, Ibis. 146, pp.50-59. Prigmore, J. (2003). The foraging behaviour of breeding Ring Ouzels (Turdus torquatus) at Glen Clunie, Scotland. MSc, University of Aberdeen. Rebecca, G.W., (2001). The contrasting status of the Ring Ouzel in 2 areas of upper Deeside, north east Scotland between 1991 and 1998, Scottish Birds. 22, pp.9-19. 29

Robinson, R.A., Green, R.A., Baillie, S.R., Peach, W.J., & Thompson, D.L., (2004). Demographic mechanisms of the population decline of the Song Thrush Turdus philomelos in Britain, Journal of Animal Ecology. 73, pp.670-682. Schmidt, K.A., Rush, S.A. & Ostfeld, R.S., (2008). Wood thrush nest success and postfledging survival across a temporal pulse of small mammal abundance in an oak forest, Journal of Animal Ecology. 77, pp.830-837. Sharrock, J.T.R. (ed), (1976). The Atlas of Breeding Birds in Britain and Ireland. Calton, Poyser. Sim, I.M.W., Burfield, I.J., Grant, M.C., Pearce-Higgins, J.W., & Brooke, M.deL., (2007). The role of habitat occupancy in determining breeding site occupancy in a declining Ring Ouzel Turdus torquatus population, Ibis. 149, pp.374-385. Sim, I.M.W., Rebecca, G.W., Ludwig, S.C., Grant, S.C., & Reid, J.M., (2011). Characterizing demographic variation and contributions to population growth rate in a declining population, Journal of Animal Ecology. 80, pp.159-170. Sim, I., Rollie, C., Arthur, D., Benn, S.,Booker, H., Fairbrother, V., Green, M., Hutchinson, K., Ludwig, S., Nicoll, M., Poxton, I., Rebecca, G., Smith, L., Stanbury, A., & Wilson, P., (2010). The decline of the ring ouzel in Britain, British Birds. 103, pp.229-239. Stokke, B.G., Anders, P.M., Saether, B.E., Rheinwald, G., & Gutscher, H., (2005). Weather in the breeding area and during migration affects the demography of a small long-distance passerine migrant, The Auk. 122 (2), pp.637-647. Suedkamp-Wells, K.M., Ryan, M.R., Millspaugh, J.J., Thompson III, R., & Hubbard, M.W., (2007). Survival of post fledging birds grassland birds in Missouri, The Condor. 109, pp.781-794. Tyler, S.J., & Green, M., (1994). The status and breeding ecology of Ring Ouzels Turdus torquatus in Wales with reference to soil acidity, Welsh Bird Report. 7, pp.78-89. White, J.D., Gardali, T., Thompson III, F.R. & Faaborg, J., (2005). Resource selection by juvenile Swainson s Thrushes during the postfledging survival, The Condor. 107, pp.388-401. 30

Wotton, S.R., Langston, R.H.W., & Gregory, R.D., (2002). The breeding status of Ring Ouzel Turdus torquatus in the UK in 1999, Bird Study. 49, pp.26-34. 31

Appendix I: Glen Clunie Provisioning Watch Raw Data This appendix contains all data collected during the food provisioning watches in Glen Clunie. Nest Territory Day of Season (from May 1) Brood Size Brood Age (days) Time of Watch Early/ Late Brood No. of Feeds Mean Prey Load Size (x size of bill) % of Feeds Containing Earthworms % of Feeds Containing Caterpillars % of Feeds Containing Larvae 1 19 4 10 14:45 Early 14 2.71 29 21 14 2 26 3 9 15:00 Early 11 3.00 36 27 18 3 19 4 8 10:15 Early 9 3.14 33 11 11 3 20 4 9 14:30 Early 6 2.25 17 17 0 3 55 2 9 10:30 Late 7 4.43 14 14 71 4 45 4 11 10:30 Late 10 4.89 60 50 40 5 55 4 10 15:15 Late 6 4.00 33 0 17 6 14 3 10 10:00 Early 14 2.82 79 0 0 6 16 3 12 10:00 Early 12 2.70 67 17 0 6 47 5 7 12:30 Late 8 4.17 75 38 25 6 48 5 8 11:00 Late 9 4.71 11 56 44 7 51 4 8 16:15 Late 5 1.00 20 0 0 8 70 4 10 12:45 Late 14 4.43 43 7 50 9 12 3 9 11:15 Early 7 2.25 43 14 0 9 51 4 11 11:45 Late 9 4.71 0 11 67 10 37 4 9 12:00 Early 9 3.25 89 0 11 11 65 4 7 11:00 Late 11 3.00 0 9 18 12 56 4 11 17:45 Late 8 N/A N/A N/A N/A 32

Appendix II: Glen Effock Provisioning Watch Raw Data This appendix contains all data collected during the food provisioning watches in Glen Effock. Nest Territory Day of Season (from May 1) Brood Size Brood Age (days) Time of Watch Early/ Late Brood No. of Feeds Mean Prey Load Size (x size of bill) % of Feeds Containing Earthworms % of Feeds Containing Caterpillars % of Feeds Containing Larvae 12 17 4 7 13:00 Early 8 4.33 0 38 13 13 63 4 10 11:00 Late 11 4.29 11 11 67 14 64 4 7 13:45 Late 11 4.83 8 23 38 15 21 4 8 13:30 Early 11 3.56 36 9 9 16 22 4 7 10:30 Early 10 4.67 45 9 9 17 22 4 11 15:00 Early 11 3.57 18 0 27 17 57 3 11 12:30 Late 13 4.33 30 10 60 18 29 4 11 12:30 Early 9 4.18 18 9 64 18 29 4 12 10:00 Early 16 3.54 46 23 69 19 42 4 9 12:45 Late 6 4.00 67 22 56 20 43 4 8 10:45 Late 5 3.63 50 0 38 21 43 5 7 14:00 Late 9 4.00 33 17 50 22 57 4 7 15:45 Late 13 5.50 20 20 80 33

Appendix III: Additional Research Photos Glen Clunie study area 34

Portable hide used for provisioning watches Nestling aged 12 days old during ringing 35

Provisioning adult male carrying earthworm Nestlings at 9 days old 36