See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/235963724 Nest site selection by Hen Harriers in Scotland ARTICLE in BIRD STUDY · MARCH 1998 Impact Factor: 1.11 · DOI: 10.1080/00063659809461077 CITATIONS READS 27 12 7 AUTHORS, INCLUDING: Steve Redpath University of Aberdeen 172 PUBLICATIONS 3,443 CITATIONS SEE PROFILE Available from: Steve Redpath Retrieved on: 14 January 2016 This art icle was downloaded by: [ 218.57.136.200] On: 21 March 2014, At : 17: 29 Publisher: Taylor & Francis I nform a Lt d Regist ered in England and Wales Regist ered Num ber: 1072954 Regist ered office: Mort im er House, 37- 41 Mort im er St reet , London W1T 3JH, UK Bird Study Publicat ion det ails, including inst ruct ions f or aut hors and subscript ion inf ormat ion: ht t p: / / www. t andf online. com/ loi/ t bis20 Nest site selection by Hen Harriers in Scotland S. Redpat h , M. Madders , E. Donnelly , B. Anderson , S. Thirgood , A. Mart in & D. Mcleod Published online: 29 Mar 2010. To cite this article: S. Redpat h , M. Madders , E. Donnelly , B. Anderson , S. Thirgood , A. Mart in & D. Mcleod (1998) Nest sit e select ion by Hen Harriers in Scot land, Bird St udy, 45: 1, 51-61, DOI: 10. 1080/ 00063659809461077 To link to this article: ht t p: / / dx. doi. org/ 10. 1080/ 00063659809461077 PLEASE SCROLL DOWN FOR ARTI CLE Taylor & Francis m akes every effort t o ensure t he accuracy of all t he inform at ion ( t he “ Cont ent ” ) cont ained in t he publicat ions on our plat form . However, Taylor & Francis, our agent s, and our licensors m ake no represent at ions or warrant ies what soever as t o t he accuracy, com plet eness, or suit abilit y for any purpose of t he Cont ent . Any opinions and views expressed in t his publicat ion are t he opinions and views of t he aut hors, and are not t he views of or endorsed by Taylor & Francis. The accuracy of t he Cont ent should not be relied upon and should be independent ly verified wit h prim ary sources of inform at ion. Taylor and Francis shall not be liable for any losses, act ions, claim s, proceedings, dem ands, cost s, expenses, dam ages, and ot her liabilit ies what soever or howsoever caused arising direct ly or indirect ly in connect ion wit h, in relat ion t o or arising out of t he use of t he Cont ent . This art icle m ay be used for research, t eaching, and privat e st udy purposes. Any subst ant ial or syst em at ic reproduct ion, redist ribut ion, reselling, loan, sub- licensing, syst em at ic supply, or dist ribut ion in any form t o anyone is expressly forbidden. Term s & Condit ions of access and use can be found at ht t p: / / www.t andfonline.com / page/ t erm s- and- condit ions Bird Study (1998) 45, 51–61 Nest site selection by Hen Harriers in Scotland Downloaded by [218.57.136.200] at 17:29 21 March 2014 STEVE REDPATH,1* MIKE MADDERS,2 ERIC DONNELLY,1 BRUCE ANDERSON,3 SIMON THIRGOOD,4 ANN MARTIN1 and DAVID MCLEOD5 1Institute of Terrestrial Ecology, Hill of Brathens, Banchory, Kincardineshire, Scotland AB3 4BY, UK, 2RSPB, Carnduncan, Gruinart, Isle of Islay, Argyll, Scotland PA44 7PS, UK, 3RSPB, The Neuk, Coupar Angus, Perthshire, Scotland, UK, 4The Game Conservancy Trust, Crubenmore Lodge, Newtonmore, Inverness-shire, Scotland PH20 1BE, UK, 5Department of Geography, Edinburgh University, Edinburgh, Scotland, UK The aim of this study was to examine nest site selection by Hen Harriers Circus cyaneus at two spatial scales in Scotland and to assess whether breeding success was influenced by choice of nest site. At the landscape scale, we compared availability and utilization of habitats in 610 km2 of Argyll, and at the local scale, we compared vegetation and topography at 52 harrier nests and random points within three areas of heather moorland covering a total area of 462 km2. At both scales, harriers showed a clear preference for nesting in heather. Within heather moorland, harriers nested in taller heather (average height 46cm) and nearer streams than expected by chance. More nests were on northwest-facing slopes than expected by chance. We found no evidence that breeding success was influenced by habitat or topography. Heather moorland is declining in the uplands due to overgrazing and afforestation. The association of harriers with heather suggests that their future may become increasingly dependent on moorland, where heather is maintained for grouse. T he distribution of nesting raptors is influenced by nest site and food availability.1 Provided suitable nest sites are available, raptors can be expected to settle in areas where food is abundant. Within areas, breeding success may be significantly improved by selecting sites which minimize the risks of predation,2–4 and optimize the thermal environment.3–5 In reality, the choice of nest site may represent a compromise between these factors.4 The Hen Harrier is a raptor of open country, which nests on the ground and selectively hunts the edges between habitats.6,7 In Britain, the principal prey species by number during the breeding season is the Meadow Pipit Anthus pratensis, although a variety of other species, including Red Grouse Lagopus lagopus *Correspondence author. Email: smre@wpo.nerc.ac.uk © 1998 British Trust for Ornithology scoticus chicks are also taken.8–10 In continental Europe, North America and historically in Britain,8 harriers nest in a variety of habitats, although moorland dominated by Heather Calluna vulgaris currently appears to be the preferred breeding habitat of British Hen Harriers, with young forestry plantations used to a lesser extent.11 Of 922 nests examined by the RSPB (unpubl.), 76% were in heather moorland, of which 49% were located on moorland managed for Red Grouse. Despite the number of harrier nests located in Britain, nest site selection has yet to be quantified. Studies have indicated that Hen Harriers appear to prefer tall heather below an altitude of about 500 m and often choose nest sites close to streams in valley bottoms.8,12 In this paper we examine the distribution of Hen Harrier nests in relation to vegetation at two scales. First we consider nest selection at the landscape scale, using data collected from 52 S. Redpath et al. aerial photographs, and secondly we look at finer scale selection within heather moorland areas. Finally we consider how breeding success is influenced by choice of nest site and discuss the implications for Hen Harrier populations in Britain. extracted from a geographic information system (ARCINFO) based on the interpretation of 1:24 000 black and white aerial photographs, by the Macauley Land Use Research Institute. From these 1988 photographs, the data were captured for 50 m × 50 m tiles (0.25 ha). Originally 1501 habitat classes were identified, although we excluded unsuitable habitat such as water and aggregated remaining classes into five broad groups, corresponding with NVC (National Vegetation Classification)13 categories: heaths (1); mires, swamps, tall herbs and upland grassland (2); woodland, divided into open canopy woodland (3) and closed canopy woodland (4); and remaining habitats (5). Within the four areas, we measured the amount of each of these five habitat categories and then examined the habitat at harrier nest sites. Although harrier nests were located from METHODS Downloaded by [218.57.136.200] at 17:29 21 March 2014 Habitat at the landscape scale At a landscape level, we compared availability with utilization by nesting harriers, of various habitat types in four areas of Argyll (Fig. 1), covering a total of 610 km2 (Table 1). Within these four areas, the harriers were relatively free from persecution and had a wide range of habitats available to them. The areas were searched thoroughly for harrier nests from 1988 to 1996. Habitat data for each area were C Argyll B A Fig. 1. Map of Scotland showing location of study areas. Within the Argyll box, the four study areas are indicated. Areas A, B and C refer to areas of heather moorland where nest site details were taken. © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 Hen Harrier nest sites 53 Table 1. Size of areas where study was conducted and number of nests on each. Random points indicate number of initial random quadrats and number of additional quadrats from rank heather stands (H). The area of habitat measured indicates the size of area measured from aerial photographs. Downloaded by [218.57.136.200] at 17:29 21 March 2014 Area No. of estates Size of area No. of nests No. of random points Area of habitat measured Landscape scale Argyll – 610 km2 34 – 610 km2 Local scale A B C 1 2 3 178 km2 120 km2 164 km2 11 19 22 30 +30H 60 +60H 35 164 km2 48 km2 21 km2 1988 to 1996, to ensure that nest data were independent, we only used data for the year when most nests were present in each of the four areas. For analysis the data from the four areas were combined. Habitat at the local scale At a finer scale, we compared habitat availability versus utilization on six upland estates in three areas (A, B & C) in Scotland between 1993 and 1995 (Fig. 1). Area A consisted of one large estate in Dumfriesshire, southwest Scotland; area B consisted of two neighbouring estates in Ayrshire, southwest Scotland; and area C comprised three neighbouring estates in Perthshire, central Scotland (Table 1). These estates were initially selected because harriers were allowed to breed freely on them in the absence of persecution, although some broods were destroyed on one of the areas in 1993. All areas were managed for Red Grouse and this meant that heather was the principal vegetation type. Active gamekeepers were employed on each area and consequently mammalian predators and crows Corvid spp. were controlled. Purple Moor Grass Molinia caerulea was abundant on area A, Bracken Pteridium aquilinum was common on areas A and B and rushes Juncus spp. were common in all three areas. To ensure that nest data were independent, we again used data from each area for the year when most nests were present. In addition, data were collected from a minimum of 30 randomly chosen points for each area. For the random points, a grid was placed over maps of the areas and coordinates obtained using a random number generator. Any points that fell in unsuitable habitat types (e.g. water or bare rock) were ignored. In order to avoid inadvertent observer bias in habitat selection, random points were located in the field by walking to where the observer considered the point to be from the map, then taking a bearing due north and walking 100 m. At each of the nests and random points, data on vegetation and topographic features were obtained during the autumn and early winter. Vegetation data were collected within a 2-m quadrat, placed over the nest. Within quadrats, species presence and percentage cover were determined. Vegetation composition in nest and random sites was compared by classifying the vegetation into NVC categories using TABLEFIT.14 Vegetation height was measured at five 10-cm intervals on the diagonal in from each corner. This gave a total of 20 measures and avoided the actual nest cup. In addition to vegetation, the following measures were obtained: 1. Distance to the nearest stream; for consistency, only streams that were marked on 1:25 000 OS maps were considered. 2. Angle of slope; measured by eye on a scale of 1 to 5 where 1 = flat, 2 = < 20°, 3 = 20–40°, 4 = 40–60° and 5 = > 60°. 3. Direction of slope (0–360°); for statistical analysis, directions were grouped into four categories (1 = 1–90°, 2 = 91–180°, 3 = 181–270° and 4 = 271–360°). 4. Height above sea-level (m). For two of the nests in area B, the vegetation © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 54 S. Redpath et al. was burnt before measurements were taken, so for these only topographic measures were taken. Downloaded by [218.57.136.200] at 17:29 21 March 2014 Potentially suitable sites As harriers invariably nested in tall heather (see Results), we increased the number of random points in equivalent habitat in order to compare nest sites with randomly chosen suitable habitat. On two of the estates (A & B), we selected a further 30 tall heather stands (greater than or equal to the minimum height of nest site vegetation) from random points. Points were selected as before, although this time observers visited the stand of tall heather nearest to the random point and sampled the approximate centre. On one estate (A), the distribution of rank or degenerate heather (i.e. tall heather with the canopy opening up) was mapped from 1:24 000 black and white aerial photographs taken in 1988. Using a stereoscope, a 1-ha grid was laid over the photographs and the vegetation type and percentage cover within each hectare estimated. Validation in the field (groundtruthing) revealed that it was relatively straightforward to determine the presence of rank heather. The distribution of rank heather was measured as the number of hectares containing that habitat in each km2. Prey abundance Meadow Pipit abundance was estimated in 18 1-km2 sites chosen at random over estate A. Random sites chosen in enclosed pasture or woodland were excluded. Two parallel 1-km transects, set 500 m apart, were walked within each site. Surveying was conducted in early summer, between 0500 and 0900 hours on calm days with good visibility.15 For each transect, abundance was simply measured as the number of pipits counted within a 200-m strip of the transect. Abundance within each km2 was taken as the mean of the two transects. Abundance was compared to the amount of heather and rank heather within each count area. between nesting habitat and breeding success, we compared the number of young fledged (measured as the number of young over 25 days old at nests) with the dominant vegetation at the nest site. Again we repeated this analysis at two scales: first we utilized data collected from a total of 395 harrier nests over nine years (1988–1996) from Argyll. These data were divided into those from the island of Islay, where there are no foxes, and those from mainland Argyll, where foxes are present and only lightly controlled. Secondly, we compared breeding success with habitat and environmental measures at nests on the heather moorland study areas. RESULTS Landscape scale At the two scales in which we considered site selection, Hen Harriers showed strong preference for heath vegetation. In Argyll, there was a highly significant association between nest location and habitat, with more nests located in heaths and open canopy woodland than expected by chance (Table 2) (χ2 = 39.7, 4 df, P < 0.001). In Argyll the open canopy woodland referred to young conifer plantations and, within this habitat, 14 (87%) of the nests were located in heather-dominant vegetation. In the closed canopy woodland, the nest occurred in a heather ride within the wood. Table 2. Nest site location in four habitat types in 610 km2 of Argyll. Number of expected nests is derived from the relative area of that habitat. Number of nests Habitat Area (ha) Observed Heaths Mires/swamps/ upland grassland Open canopy woodland Closed canopy woodland Rest 10 244 17 272 15 2 5.7 9.6 13 711 16 7.6 11 353 1 6.3 8 386 0 4.8 Total area 60 965 Breeding success To see whether or not there was any association © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 Expected Hen Harrier nest sites 55 Moor A Proportion 1 0.8 0.6 0.4 0.2 U20 U16 U5 U4 U2 SD17 M25 M23 M15 H21 H18 H11 H10 H1 0 NVC category Moor B Proportion Downloaded by [218.57.136.200] at 17:29 21 March 2014 1 0.8 0.6 0.4 0.2 U20 U16 U5 S11 M25 U2 M25 M23 M23 M17 M16 M15 M6 H21 H18 H13 H10 H9 H1 0 NVC category Moor C Proportion 1 0.8 0.6 0.4 0.2 U6 U5 S11 M20 M15 H21 H18 H13 H11 H10 H9 H1 0 NVC category Fig. 2. The proportion of (■) nest and (❏) random sites in various NVC (National Vegetation Classification) categories for areas A, B and C. Local scale Within heather moorland, a comparison of NVC categories (Table 3) used for nesting and those selected at random in the three areas showed that nests fell in eight of the 22 categories selected from the random quadrats (Fig. 2). Of the 50 nests where habitat could be measured, 94% were in Calluna-dominant vegetation, with the remaining 6% in rushes Juncus spp. (all in area B). On all areas, the most widely used NVC category was H1 Calluna–Festuca ovina heath. For statistical comparison, NVC categories were grouped into four broad types: H, heaths; M, mires; S, swamps and tall-herb fens; U, upland grassland. There were significant differences in each of the three areas, with more nests located in heaths than expected by chance, and fewer nests in the other three categories (Table 4). For a comparison of vegetation height, only those quadrats which fell in NVC categories © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 56 S. Redpath et al. Table 3. Definitions of NVC categories determined from 2 × 2 m quadrats using the TABLEFIT program.14 Downloaded by [218.57.136.200] at 17:29 21 March 2014 NVC Category Habitat definition H1 H9 H10 H11 H13 H18 H21 M6 M15 M16 M17 M20 M23 M25 S11 SD17 U2 U4 U5 U6 U16 U20 Calluna vulgaris–Festuca ovina heath Calluna vulgaris–Deschampsia flexuosa heath Calluna vulgaris–Erica cinerea heath Calluna vulgaris–Carex arenaria heath Calluna vulgaris–Cladonia arbuscula heath Vaccinium myrtillus–Deschampsia flexuosa heath Calluna vulgaris–Vaccinium myrtillus – Sphagnum capillifolum heath Carex echinata–Sphagnum recurvum / auriculatum Scirpus caespitosus–Erica tetralix Erica tetralix–Sphagnum compactum Scirpus caespitosus–Eriophorum vaginatum Eriophorum vaginatum blanket/raised bog Juncus effusus/acutiflorus–Galium palustre Molinia caerulea–Potentilla erecta mire Carex vesicaria swamp Potentilla anserina–Carex nigra Deschampsia flexuosa Festuca ovina–Agrostis capillaris–Galium saxatile Nardus strict –Galium saxatile Juncus squarrosus–Festuca ovina Luzula sylvatica–Vaccinium myrtillus Pteridium aquilinum–Galium saxatile which were utilized for nesting were used. Combined data from the three areas indicated that nests were located in significantly taller Calluna vegetation than the random points (Fig. 3; nests: 46.0 ± 1.3 cm, random: 27.9 ± 2.0 cm; t-test: t = 7.6, df = 80, P < 0.001). When comparing topographic and other features between nests and random sites, only potentially suitable vegetation types were selected. Therefore only random points (including the extra randomly located heather stands) which consisted of the NVC categories which actually held nests and were > 25 cm tall were used. This left a total of 78 random points between the three areas. Random points did not differ between the three areas in terms of Table 4. Distribution of observed frequency of harrier nests and random points in four NVC categories. Two nest sites were excluded from area B because the vegetation had been burnt between nesting and vegetation surveying. Differences were significant for each area. A: G = 19.1, P < 0.001, B: G = 25.7, P < 0.001, C: G = 8.7, P < 0.05. Area A NVC category B C Nest Random Nest Random Nest Random Heaths Mires Swamps + tall-herbs Upland grassland 10 1 9 13 12 4 13 27 21 1 26 4 0 1 1 1 0 1 0 7 0 19 0 4 Totals 11 30 17 60 22 35 © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 Hen Harrier nest sites 57 50 45 40 30 25 20 15 10 5 90–100 80–90 70–80 60–70 50–60 40–50 30–40 20–30 10–20 0 <10 Downloaded by [218.57.136.200] at 17:29 21 March 2014 Occurrence (%) 35 Vegetation height (cm) Fig. 3. Distribution of vegetation heights at (■) nest and (❏) random sites on three areas combined. Random sites falling in NVC categories which were not used for nesting were excluded. distance to stream (ANOVA F = 2.7, df = 2, ns), angle of slope (Kruskal–Wallis H = 2.6, df = 2, ns) or direction of slope (G = 6.1, df = 6, ns) and therefore data were combined for the areas. The majority of nests were within 50 m of a stream and, overall, nests were closer to streams than the subset of random points (Fig. 4; nests 93 ± 18 m, random 154 ± 15 m, t = –2.7, P < 0.01). Within the random data set there was no correlation between the distance to a stream and vegetation height (r = –0.16, N = 78, P = 0.16). Comparing nests and random points, there was no difference in the angle of slope (nests = 52, random = 78, Mann–Whitney U = 3680, ns), although the direction that the slopes faced differed between nests and random (nests = 49, random = 65, G = 13.22, P < 0.01), with more nests located in quadrant 4 (i.e. northwest facing) than expected from random and fewer in the other three quadrants. Three nests and thirteen of the random sites occurred on flat ground, so there was no measure of slope direction. Comparing nest sites on northwest-facing slopes with the other quadrants combined, neither slope angle (Mann–Whitney U = 478, P = 0.9), vegetation height (t = 0.41, P = 0.7), nor distance to stream (t = 0.66, P = 0.5) varied with slope direction. Nests were located between 190 m and 520 m above sea-level. The three areas varied in altitude (random points ANOVA F2,75 = 6.5, P < 0.01), so nests were compared to random on the areas separately. There were no differences on two of the areas (A: nests 305 ± 30 m, random 309 ± 20 m, U = 200, P = 0.9. C: nests 390 ± 12 m, random 386 ± 11 m, U = 439, P = 0.8). On area B nests were higher than expected by chance (nests 377 ± 8 m, random 352 ± 8 m, U = 658, P < 0.05). Distribution of heather and passerines From the random quadrats, tall heather (over 40 cm) was estimated to form 10% of area A, 5% of area B and 26% of area C. Within area A, the distribution of rank heather was highly © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 58 S. Redpath et al. 60 Occurrence (%) 50 40 30 10 550–600 500–550 450–500 400–450 350–400 300–350 250–300 200–250 150–200 100–150 50–100 0 0–50 Downloaded by [218.57.136.200] at 17:29 21 March 2014 20 Distance to stream (m) Fig. 4. Distribution of distances from (■) nest and (❏) random sites to the nearest stream. Random sites falling in NVC categories and vegetation heights not used for nesting were excluded. aggregated (Fig. 5). Comparing heather cover in km2 with and without harrier nests revealed that nests were placed in areas where there was a greater percentage cover of rank heather (Fig. 5. Mann–Whitney U = 906.5, n1 = 39, n2 = 11, P < 0.001). Harriers appeared to select areas of the moor for nesting where there was most rank heather. Passerine abundance varied from 16.5 to 31.5 birds km–1 (mean = 23.8 + 1.0, n = 18). There was no association between Meadow Pipit abundance and either rank heather (rp = –0.05, n = 18, ns) or total heather cover (rp = –0.08, n = 18, ns). Breeding success Argyll harrier nest vegetation was classified as either heather (N = 343) or non-heather (mostly in grass or rushes N = 52). There was © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 no difference in the proportion of successful nests between the two habitat types, either on Islay (G = 0.15, 1 df, ns) or on the mainland (G = 0.20, 1 df, ns). Similarly, for all nesting attempts, there were no differences in the number of young produced, either on Islay (heather: N = 205, mean = 2.42 ± 0.1; other: N = 32, mean = 2.72 + 0.3, t = –0.93; P = 0.36) and on the mainland (heather: N = 138, mean = 2.48 + 0.15; other: N = 20, mean = 2.5 ± 0.38; t = –0.05, P = 0.96). Harriers did not produce more young from nests in heather-dominant vegetation. At a fine scale we also found no relationship between breeding success and any of the measured variables. A multiple regression model was used, excluding data from the 11 nests on one estate where human interference was suspected. The independent variables included in the model were altitude, distance to stream and vegetation height. Hen Harrier nest sites 59 114 12 8 6 1 1 3 1 1 1 2 1 4 2 30 28 26 24 22 20 18 16 14 12 10 8 6 4 2 0 0 Downloaded by [218.57.136.200] at 17:29 21 March 2014 Frequency 10 Cover (%) Fig. 5. Distribution of rank heather, within area A, as measured by the number of hectares in each km2 containing rank heather (❏). (■) The km2 containing harrier nests and the numbers above the columns indicate the number of harriers nesting in each category. Overall these factors accounted for less than 1% of the variation in breeding success (F3,37 = 0.85, P = 0.48). As altitude varied between areas, the analysis was then conducted for each separately, although it still revealed no significant relationships (Area A: F3,6 = 1.8, P = 0.25; Area B: F3,8 = 0.72, P = 0.56; Area C: F3,15 = 0.43, P = 0.74). As there appeared to be some nest site selection in the direction the slope was facing, an analysis of variance was used to see whether breeding success varied between the four quadrants. There were no significant differences (F3,36 = 0.13, P = 0.94). DISCUSSION The Hen harriers in this study preferentially nested in heather and this was apparent at the landscape scale in Argyll and at the local scale within managed heather moorland. In continental Europe and North America, Hen Harriers nest in a variety of tall ground or scrub vegetation types,4,8,16–18 and within Britain nest sites have been located in forest brash, rushes, Bracken, willow Salix spp. and Purple Moor Grass.8,13 On the three areas where nest habitat was examined in detail, rushes and Bracken were present on all areas and Purple Moor Grass was abundant on two of the areas (A & B). Despite this, only three of 52 nests located were in vegetation other than tall heather: two in rushes and one in a mixture of rushes and Purple Moor Grass. Within heather moorland, harriers chose nest sites which were nearer streams than expected by chance. This was not because the vegetation was taller there, although it may have been due to some unmeasured variable, such as the degree of shelter from the wind or the moisture © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 Downloaded by [218.57.136.200] at 17:29 21 March 2014 60 S. Redpath et al. content of the soil. There was also some selection for northwest-facing slopes. This preference may result from a need to place nests with some shelter from direct sunlight or from prevailing wind and rain. If harriers chose nest sites adaptively then they should choose sites which confer an advantage in terms of improved breeding success. Simmons & Smith4 presented some evidence to suggest that nest site selection influenced the breeding success of Northern Harriers in Canada. However, we found no evidence that breeding success was improved by any of the measured habitat variables. In Argyll, Hen Harriers nesting in heather did not fledge more offspring than those in grass or rushes. Similarly, on the three areas of heather moorland, fledging success did not appear to be related to any of the measured variables. We can think of three possible reasons why harriers should be so selective in their choice of site and yet show no apparent benefit. First, site selection may represent an expression of past pressures on the birds, when their predators may have been more abundant and site selection was more critical in determining successful breeding. Second, an advantage may be noticeable only in years of extreme weather conditions or in areas of high predator abundance. Third, the preference for heather in Britain may be cultural. Harriers were pushed to the point of extinction by gamekeepers at the start of this century, with birds breeding only in the outer islands off the west and north coasts.8 If these birds were heather breeders it is possible that their offspring only recognize that habitat as suitable breeding habitat. The latter two hypotheses could be tested by examining breeding success in relation to weather and predator abundance and examining tagged harriers to see if there is an inherited component to site selection. On one area examined in detail, nesting habitat was highly aggregated, so harriers were restricted to a relatively small area of moorland. In this area, the harriers’ main prey was the Meadow Pipit (unpubl. data) and this species was relatively abundant throughout the estate and was not associated specifically with heather. It is unlikely therefore that Meadow Pipits influenced the distribution of nesting harriers within areas. Surveys of upland birds over larger geographical areas are ambiguous © 1998 British Trust for Ornithology, Bird Study, 45, 51–61 as to the relationship between pipit abundance and heather. Brown & Stillman19 found no evidence of an association in northeast Scotland, although positive relationships have been found in Wales20 and Caithness and Sutherland.21 Whether harriers have selected heather as a nesting site because of the availability of grouse is unclear. Male harriers feed their females during courtship and incubation, and they generally feed on small prey and not on adult grouse.8 Therefore males can be expected to settle in relation to the availability of small prey rather than grouse. However, grouse can be an important component of female diet in winter22 and thus grouse density could influence nest site selection by females. Future implications for Hen Harrier populations The close association between breeding Hen Harriers and heather in Scotland means that they are vulnerable to land use changes due to grazing and forestry planting and to persecution from gamekeepers. Heather moorland, and hence harrier breeding habitat, is declining throughout the British uplands, either through overgrazing or through afforestation.23–25 Harriers do utilize heather in young forestry plantations, and the change from heather moorland or upland grassland to forestry can initially be beneficial. This benefit is only short term however, as harriers tend to avoid forests once the canopy cover closes.8 They have been recorded nesting in second rotation plantations,26 although nest records kept by the RSPB suggest that this is extremely rare (Etheridge, pers. comm.). Since 1988, changes in financial incentives have meant that the amount of new forestry in Britain has decreased substantially27 and, as the amount of this habitat declines, the relative importance of managed heather moorland to harriers is likely to increase. Large areas of heather moorland are currently maintained for Red Grouse, and the association between harriers and heather brings these raptors into conflict with grouse-shooting interests. However, the reliance of harriers on heather suggests that the future of populations in mainland Britain may increasingly be dependent on those who manage grouse moors for sporting purposes. Hen Harrier nest sites ACKNOWLEDGEMENTS Downloaded by [218.57.136.200] at 17:29 21 March 2014 We are grateful to the land owners for allowing us access to their land and in particular to Buccleuch Estates for their logistic support. Chris Hill, Ian Miller, Kate Redpath and Bob Stakim helped with aspects of the fieldwork. Peter Carey helped with vegetation analysis. This work was funded by the Game Conservancy Trust, Institute of Terrestrial Ecology, Buccleuch Estates Ltd, The Royal Society for the Protection of Birds, Scottish Natural Heritage, The Joint Nature Conservation Committee, The Scottish Research Trust and an anonymous donor. REFERENCES 1. Newton, I. (1979) The Population Ecology of Raptors. Poyser, Berkhamsted. 2. Mearns, R. & Newton, I. (1988) Factors affecting breeding success of Peregrines in south Scotland. J. Anim. Ecol., 57, 903–916. 3. Skutch, A.F. (1976) Parent Birds and Their Young. University of Texas. 4. Simmons, R. & Smith, P.C. (1984) Do northern harriers (Circus cyaneus) choose nest sites adaptively? Can. J. Zool., 63, 494–498. 5. Mosher, J.A. & White, C.M. (1976) Directional exposure of golden eagle nests. Can. Field. Nat., 90, 356–359. 6. Schipper, W.J.A. (1977) Hunting in three European harriers (Circus) during the breeding season. Ardea, 65, 53–72 7. Redpath, S.M. (1992) Behavioural interactions between hen harriers and their moorland prey. Ornis Scand., 23, 73–80. 8 Watson, D. (1977) The Hen Harrier. Poyser, Berkhamsted. 9. Picozzi, N. (1978) Dispersion, breeding and prey of the hen harrier Circus cyaneus in Glen Dye, Kincardineshire. Ibis, 120, 498–509. 10. Redpath, S.M. (1991) The impact of hen harriers on red grouse breeding success. J. Appl. Ecol., 28, 659–671. 11. Bibby, C.J. & Etheridge, B. (1993) Status of the hen harrier Circus cyaneus in Scotland in 1988–89. Bird Study, 40, 1–11. 12. Balfour, E. (1962) The nest and eggs of the hen harrier in Orkney. Bird Notes, 30, 69–73, 145–153. 13. Rodwell, J. (ed) (1991) British Plant Communities. 61 Vol. 2. Mires and Heaths. Cambridge University Press, Cambridge. 14. Hill, M.O. (1993) TABLEFIT version 0.0, for identification of vegetation types. Institute of Terrestrial Ecology, Huntingdon. 15. Thirgood, S.J., Leckie, F.M. & Redpath, S.M. (1995) Diurnal and seasonal variation in line transect counts of moorland passerines. Bird Study, 42, 257–259. 16. Schipper, W.J.A. (1978) A comparison in breeding ecology in three european raptors. Ardea, 66, 77–102. 17. Hamerstrom, F. & Koppeny, M. (1981) Harrier nest site vegetation. Raptor Res., 15, 86–88. 18. Cormier, J.P. (1984) Le rôle de la végetétion dans l’emplacement des sites de réproduction chez Circus cyaneus L. et Circus pygargus L. dans des secteurs de reboisement en conifères. Rev. Ecol. (Terre Vie), 39, 447–457. 19. Brown, A.F. & Stillman, R.A. (1993) Bird–habitat associations in the eastern highlands of Scotland. J. Appl. Ecol., 30, 31–43. 20. Tyler, S.J. (1988) Birds and bracken in Wales. Bracken in Wales, Annex 5. Nature Conservancy Council, Bangor. 21. Stroud, D.A., Reed, T.M., Pienkowski, M.W. & Lindsay, R.A. (1987) In Birds, Bogs and Forestry: the Peatlands of Caithness and Sutherland (eds D.A. Ratcliffe & P.H. Oswald), pp. 121. Nature Conservancy Council, Peterborough. 22. Marquiss, M. (1980) Habitat and diet of male and female hen harriers in Scotland in winter. Br. Birds 73, 555–560. 23. Anderson, N.P. & Yalden, D.W. (1981) Increased sheep numbers and the loss of heather moorland in the Peak District, England. Biol. Conserv., 20, 195–213. 24. Bibby, C.J. (1988) Impacts of agriculture on upland birds. In Ecological Change in the Uplands (eds M.B. Usher & D.B.A. Thompson), pp. 223–236. British Ecological Society Special Publication No. 7, Blackwell Science, Oxford. 25. Thompson, D.B.A., MacDonald, A.J., Marsden, J.H. & Galbraith, C.A. (1995) Upland heather moorland in Great Britain: a review of international importance, vegetation change and some objectives for nature conservation. Biol. Conserv., 71, 163–178. 26. Petty, S.J. & Anderson, D. (1986) Breeding by hen harriers Circus cyaneus on restocked sites in upland forests. Bird Study, 33, 177–178. 27. Avery, M. & Leslie, R. (1990) Birds and Forestry. T & AD Poyser, London. (MS received 8 November 1996, revised MS accepted 20 June 1997) © 1998 British Trust for Ornithology, Bird Study, 45, 51–61