Holyrood Park Kestrel survey results for 2007 – Graham Checkley 25/11/2008
Introduction
The purpose of this document is to provide an account of the status of the Kestrel (Falco tinnunculus) within Holyrood Park, Edinburgh during 2007. As such it provides information on the distribution, occupancy and breeding success of the species based on the methods documented in ‘Bird Monitoring Methods’ (Gilbert, Gibbons and Evans, 1998), but with the addition of casual Kestrel records acquired from Historic Scotland Ranger Service patrols.
Full length electronic versions of this document, including distribution maps, tables, references and appendices may be obtained from the author.
Background
The Kestrel is the most common raptor breeding in the UK, able to adapt to most habitats, including city centres. It is scarce only in north-west Scotland, the Western Isles and Shetland (Gilbert, Gibbons and Evans, 1998), but there has been a general decline in Kestrel numbers between the mid-1970s and the mid 1980s, possibly due to the effects of agricultural intensification on farmland habitat and small mammal populations (BTO, 2005). While numbers appear to have stabilised subsequently, the most recent analysis of the BTO Nest Recording Scheme data suggest that brood sizes have declined down to an average of three chicks being reared from a previous average of between four and five. As a result the Kestrel has now been added to the BTO Nest Recording Scheme Concern List (BTO, 2006).
The Historic Scotland Ranger Service (HSRS) established a log of casual wildlife sightings in January 2000 and since then has regularly recorded the presence of Kestrel within the Park, the observed locations being detailed in Appendix One of the full length version. Birds have been mainly noted adjacent to the west facing Park crags and associated grassy areas, with strong suggestions of breeding behaviour during 2000 and 2001 at Salisbury Crags (HSRS, 2006). Prior to this one or two pairs were noted as regularly breeding on Salisbury Crags and below Arthur’s Seat, but the number of breeding pairs were considered to have fallen in recent years from a maximum of six in the 1960’s (SWT, 1993).
As the 1993 Breeding Bird Census represents the last systematic study of the species within the Park it was decided, following discussions with Natalie Taylor of HSRS, to carry out a distribution, occupancy and breeding success survey of the population in 2007.
Aims
To determine the distribution, occupancy and breeding success of Kestrel within Holyrood Park during the summer 2007 season.
Objectives
To undertake a survey of Kestrel breeding season population within Holyrood Park.
To target casual records for Kestrel from HSRS patrols in order to obtain information on distribution and habitat use.
Methodology
Introduction
The approach adopted was based on the Kestrel survey methodology for breeding season population, as documented in ‘Bird Monitoring Methods’ (Gilbert, Gibbons and Evans, 1998), but with the additional goal of acquiring casual Kestrel records from HSRS patrols.
The area surveyed, namely Holyrood Park, is 2.6km² in extent, and as such falls beneath the 50km² limit defined as the area needed to accurately determine the population density for the species (Gilbert, Gibbons and Evans, 1998). As a result the survey concentrated on monitoring occupancy, but also sought to acquire additional information on species distribution and Park usage through the analysis of any casual records obtained by HSRS during the survey period.
Prior knowledge acquisition
The observers familiarised themselves with the description and illustrations of Kestrel in ‘Collins Bird Guide’ (Svensson and Grant, 1999) and ‘RSPB Birds of Britain and Europe’ (Hume, 2002). As a matter of principle they were also encouraged to consult a variety of field guides, as each could provide an additional insight to the species, or provide guidance more appropriate to the reader’s level of experience.
The description and illustrations of the potential confusion species Sparrowhawk (Accipiter nisus), Peregrine (Falco peregrinus), Merlin (Falco columbarius) and Hobby (Falco subbuteo) were also noted. Sparrowhawk are resident in the Park and believed to be breeding, while Peregrine are a regular but infrequent visitor. Neither Merlin nor Hobby has been recorded to date.
Timing of visits for Breeding Season Population
The surveys were scheduled for five visits between April and mid-July with at least a week between visits, the recommended date ranges being shown in the full length version of this document. Recommended survey times were between 0700 and 1900 BST.
Survey Areas
The transect maps in Appendix One of the full length version show the major locations and transect routes followed during the first three visits. In order to make the survey manageable in a day by a single observer the route to be followed was split into two sections, transect route one and transect route two, which were visited alternately, morning and afternoon, in order to avoid survey bias.
The transect routes followed during visits four and five were determined by producing a consolidated map of recorded presence from the first three surveys, adding any casual records from HSRS patrols, then linking the grid reference points for the observations using the network of footpaths within the Park (see observation maps and transect maps in Appendix Two of the full length version). Once these points and distances were established then time was allocated for static observations concentrating on proving the occupancy or otherwise of the potential nest sites (see data, observation maps and transect maps in Appendix Two of the full length version).
Exclusions
Areas without trees and crags below three metres in height were excluded from the transect routes followed during the first three visits, as these are not considered suitable as nest sites for Kestrels (Gilbert, Gibbons and Evans, 1998). Information on Kestrel activity away from these areas was provided by casual Kestrel records from HSRS patrols.
Survey equipment
The following equipment was used by the surveyor:
Transect maps (See Appendix One and Two of the full length version)
Pencil and eraser
Binoculars (Swarovski EL 8.5 x 42)
Survey details recorded
Date
Time
Field observer
Mapping of any areas which were considered to receive insufficient coverage with reasons noted
Weather conditions (8ths of cloud cover, estimated temperature and estimated wind-speed)
Survey data collected
Location, time, count, behaviour and flight direction (if any) of Kestrels seen were recorded on the survey map.
Particular emphasis was placed on recording behaviour which included calling and fluttering-wing flight (flight level with shallow wing-beats in comparison with its normal rather looser style), both of which may be common near the nest during the April – May pre-laying period.
Other general signs of the presence of birds of prey may include young calling (squeaking), pellets and piles of feathers or vole fur, but the presence of Kestrel cannot be assumed from such observations in the field.
See Appendix One of the full length version for an example recording form.
Counting methods
The counting methods for each transect route are detailed in the full length version of this document.
It should be noted that the Kestrel is a protected species within the terms of the Wildlife and Countryside Act 1981 and as a result it is an offence to intentionally kill, injure or take a bird or their eggs or nests (JNCC, 2007). While it does not receive the legal protection of a Schedule 1 bird with regard to disturbance at nest or of their dependent young, best efforts must be made at all times to avoid any disturbance.
Survey Conditions
‘Bird Monitoring Methods’ (Gilbert, Gibbons and Evans, 1998) states that the survey should not be undertaken in poor weather, therefore survey conditions should be dry, with wind speeds less than 20 mph, but with no criteria for temperature or cloud cover.
Analysis methods to be employed
Breeding is confirmed if:
a nest containing eggs or young is observed
an adult is seen carrying food for the young
a used nest or eggshells are found (occupied or laid within the period of the survey)
recently fledged young are found.
Breeding is considered probable if:
agitated behaviour or anxiety calls are given by the adults
a pair of Kestrels is seen in suitable nesting habitat in the breeding season
a permanent territory is presumed through the registration of territorial behaviour on at least two different days separated by more than a week
courtship and display behaviour are witnessed.
Breeding is considered possible if:
a Kestrel is observed in April and/or May in possible nesting habitat
The minimum number of breeding pairs of Kestrels should be reported as the number of confirmed pairs; and the maximum number as the confirmed + probable + possible breeding pairs.
Territorial behaviour in Kestrel is typified by overt aggression by males with other males near the nest. Female aggression varies with the stage of the reproductive cycle and after hatching to the brood size (Wiklund and Village, 1992). When territory is threatened Kestrels may fly under the intruder while fanning their tails, “shivering” their wings and slowly rising under the intruding bird (Animal Diversity Web, 2007).
Courtship behaviour in Kestrel is typified by slow “shivering” flight, or by paired flight in which both sexes may fly quickly together at great heights. From the beginning of courtship until egg laying males hunt for the females and bring them prey as gifts (Animal Diversity Web, 2007).
Survey results
The full results of each survey, along with any casual Kestrel records from HSRS patrols are shown in the full length version of this document.
Discussion
Discussion of methods
Fixed survey schedule
The surveys were scheduled for five visits between April and mid-July with at least a week between visits. All but one of the surveys were completed within the recommended date range, it having to be rescheduled due to work and holiday commitments. All surveys were separated by at least a week between visits.
Survey times
Recommended survey times were between 0700 and 1900 BST. All surveys were completed within these times.
Counting methods
All transects were completed within the allotted time.
Data mapping
Analysis was facilitated by mapping the survey records onto a 1:10,000 contour map of Holyrood Park. Record data was mapped using the Ordinance Survey co-ordinate system supplied as part of the Historic Scotland implementation of the ESRI Arcmap application (version 8.3, build 800). Maps were produced as Windows bitmap files by using the ESRI Map Explorer application (version 2.0, Service Pack 1).
Discussion of 2007 fixed survey results – evidence of breeding
The 23 2007 fixed survey results were analyzed through the use of the rules shown in the section ‘Analysis methods to be employed’. They show that no more than one bird was seen at any one time during the surveys. No evidence was seen of nesting or territorial behaviour. However both male and female birds were identified during the survey, and the identification of a noticeably ragged bird on the 2nd of July might suggest the presence of a third individual.
From this survey data the minimum number of breeding pairs of Kestrels should be reported as the number of confirmed pairs (zero); and the maximum number as the confirmed (zero) + probable (zero) + possible (one) breeding pairs.
Discussion of the addition of 2007 casual records to derived breeding status
The addition of the 2007 casual records provided a further 130 records to this analysis. The same derived breeding status rules were applied as previously.
The derived breeding status shifts from possible to probable due to the observation of territorial behaviour between two male birds on the 4th and 17th of June. Also, although no evidence was found of nesting behaviour, juvenile birds and groups of three or more birds were observed during the period July to September.
This suggests that although Kestrels may not be using Holyrood Park as a nesting site two pairs may be using it as part of their territory. The presence of juveniles further indicates that at least one pair in the vicinity have been successful in breeding.
As a result, from the full survey data, the minimum number of breeding pairs of Kestrels should be reported as the number of confirmed pairs (zero); and the maximum number as the confirmed (zero) + probable (two) + possible (zero) breeding pairs.
Discussion of Kestrel distribution in Holyrood Park during 2007
Kestrels were seen in 94 of the 255 one hundred metre squares that comprise the Park. 16 immediately adjacent occupied squares occupy the area immediately west of Arthur’s Seat, and of these 13 show more than one sighting.
The areas most sparsely populated are the open water and woodland areas in the south-east corner of Duddingston / Bawsinch, areas of amenity grassland in the Galloping Glen & the Parade Ground, and most of Whinny Hill.
Discussion of Kestrel occupancy in Holyrood Park during 2007
Kestrels were seen in Holyrood Park during every month of 2007, but with an increased number of observations during the March to September period. While the number of casual records will depend on the amount of Ranger time spent in the field, the March, April and June figures may represent a real rise in Kestrel activity.
Conclusions
Kestrels are present throughout the year in Holyrood Park but may no longer nest within its boundary.
Analysis of the 2007 survey data plus casual records suggests that two pairs of birds may be using the Park as part of their territory, and that at least one pair in the vicinity had success in breeding.
This represents a marked decline from the maximum of 6 pairs in the 1960’s, and a change in usage from the early 1990’s when two pairs were noted as regularly breeding on Salisbury Crags (SWT, 1993).
The fixed survey on its own appears to understate the breeding status of Kestrel in Holyrood Park. As a result it is recommended that casual records continue to be collected during subsequent years.
The lack of change of numbers, albeit with a change of usage suggest that a full survey be repeated in ten years time.
