Prey A7ailability Influences Habitat Tolerance: an Explanation for The

FORUM is a lighter channel of communication between readers and contributors; it aims to stimulate discussion and debate, particularly by presenting new ideas and by suggesting alternative interpretations to the more formal research papers published in ECOGRAPHY and elsewhere. A lighter prose is encouraged and no summary is required. Contributions should be concise and to the point, with a relatively short bibliography. Formal research papers, however short, will not be considered.

Prey a7ailability inﬂuences habitat tolerance: an explanation for the rarity of peregrine falcons in the tropics

Andrew R. Jenkins ([email protected]) and Philip A. R. Hockey, Percy FitzPatrick Inst., Uni7. of Cape Town, Rondebosch 7701 South Africa.

The density and productivity of peregrine falcon Falco peregri- peregrine populations in their respective tropical ranges nus populations correlate positively with distance from the Equa- (Cade 1982). The pervading rarity of the peregrine tor, while habitat specificity increases with proximity to the Equator. Low peregrine densities in the tropics may be a result group in the tropics and subtropics may result from of competition with similar congeners (e.g. the lanner falcon F. competition with similar congeners (e.g. lanner falcon biarmicus in Africa), which replace them in many areas. Alterna- F. biarmicus in Africa – Tarboton and Allan 1984, and tively, tropical peregrines may be limited by resource deficiencies prairie falcon F. mexicanus in North America – Porter that do not affect their close relatives. Data from peregrine and lanner populations in South Africa support the resource defi- and White 1973), or it may be symptomatic of generally ciency hypothesis, and there is no evidence to suggest direct low resource availability in tropical environments competition between the two species. In areas where prey are not (Mendelsohn 1988, Jenkins 1991). spatially or temporally concentrated, or otherwise particularly vulnerable to attack, morphological and behavioural specializa- Here, recent peregrine population studies are re- tions of peregrines probably restrict them to optimal foraging viewed to examine postulated latitudinal trends in nest conditions. The relative dynamics of Arctic and temperate vs site use, density and breeding performance (Jenkins tropical prey populations is suggested as an important factor determining peregrine distribution globally. Populations of other 1991). Data from previous studies are synthesized in an widespread but particularly specialized avian predators (e.g. overall assessment of factors limiting South African osprey Pandion haliaetus) may be similarly controlled. Food peregrine populations, with emphasis on the relative limitation (in terms of a dearth of particularly vulnerable prey) importance of competition with sympatric lanners vs in the tropics has resulted in specialization and rarity in peregrines and generalization and relative abundance in similar more direct, environmental constraints. The primary congeners. question addressed is: if lanners were absent from the Afrotopics, would peregrine densities be higher, or are peregrine populations limited by resource deficiency, irrespective of the presence of lanners? This evaluation is then set in the wider context of other sympatric The peregrine falcon Falco peregrinus occupies an ex- populations of peregrines and congeners, relating tensive aggregate range over a wide variety of environ- predatory specialization to latitude and the dynamics of mental conditions (del Hoyo et al. 1994) but is prey populations. relatively unsuccessful in the tropics. Breeding peregrines are patchily distributed and uncommon in sub-Sa- haran Africa (Mendelsohn 1988, although see Thomsett 1988, Pepler et al. 1991, Hartley 1992), southern Asia Peregrine data from around the world (Cade 1982), northern Australia (Olsen and Olsen 1988a) and southern North America (Hunt et al. 1988, Nest site selectivity, density and breeding performance Porter et al. 1988), and they are absent from north-cen- data were extracted from recent (post-1980), published tral and eastern South America (McNutt et al. 1988). studies of peregrine populations. Many contemporary The very rare taita falcon F. faschiinucha (Afrotropics) studies focus on the productivity of falcon populations and orange-breasted falcon F. deiroleucus (Neotropics) affected by the use of agricultural pesticides, or on the are considered as specialized, peregrine equivalents, dynamics of re-colonisation in areas where the species with similar resource requirements (Hartley et al. 1993, had previously been extirpated. As far as possible, such Baker et al. 2000), and supplement or replace sparse studies were excluded from this analysis, and only data

ECOGRAPHY 24:3 (2001) 359 from naturally regulated populations – i.e. those be- (Table 2, Fig. 1). Although habitat availability in each lieved to be relatively free from anthropogenic influ- area generally was not documented, the lower minimum ences – were used. Some of the source publications height of cliffs used by temperate and Arctic popula- (especially those in Cade et al. 1988) document the tions is particularly suggestive of a greater tolerance of breeding history of peregrine populations both during sub-optimal nest sites (e.g. relatively widespread and after decreases associated with chemical contami- ground-nesting is restricted to high latitudes – Cade nation. In such cases, only data from the latter period 1982, Lindberg et al. 1988). were used. Studies conducted over multiple breeding Across the entire sample of studies, mean clutch size seasons were preferred to single-season surveys. How- and young fledged per successful pair were positively ever, in the interests of the geographic scope of the correlated with latitude (Tables 1 and 2, Fig. 2). These analysis, some studies of small populations over short trends were generally present within each hemisphere time periods were included, mostly from tropical and and in the New and Old World subsamples (Table 2). subtropical areas. Inter-pair distance, as a measure of breeding density, Correlation analysis was used to test the statistical was not significantly correlated with latitude globally, significance of latitudinal trends in nest site selection, but there was a significant negative relationship be- density and productivity. These trends were examined tween latitude and density in the Northern hemisphere, within the Northern and Southern hemispheres, and where the most comprehensive studies of the largest within the New and Old Worlds, as well as globally. populations have been made (Jenkins 1998). Peregrine and lanner biology was studied in tropical Overall, the largest, densest and most productive (the Soutpansberg), subtropical (Orange River) and peregrine populations occur at mid-high latitudes (e.g. temperate (Cape Peninsula) areas of South Africa. central Greenland – Mattox and Seegar 1988, and the Study areas and methods are detailed elsewhere Canadian Arctic – Court et al. 1988, Bradley et al. (Jenkins 1994, 1995, 1998). 1997). Conversely, the sparsest, least productive populations, and those most obviously limited (at least proximately) by the availability of suitable nesting habitat, occur in the subtropics and tropics (e.g. Peru and Latitude and the performance of peregrine Ecuador – McNutt et al. 1988, and northeastern South populations Africa – Tarboton and Allan 1984). Thirty-eight studies contributed data to this analysis Over most of North America, Eurasia and Africa, (Appendix). Most were from northern and temperate peregrines are sympatric with the ‘‘desert falcon’’ com- regions, and few data were available for tropical and plex of similar species (subgenus Hierofalco, comprising subtropical populations (Appendix, Table 1). Relatively the lanner, laggar falcon F. jugger, prairie falcon, saker few studies included usable data on nesting habitat use, falcon F. cherrug and gyrfalcon – Cade 1982), while in the density of breeding pairs and clutch size. Most were Australasia the black falcon F. subniger may be ecolog- extensive, late-season population surveys, with limited ically equivalent to members of this group (Czechura information on territory occupancy and the actual and Debus 1985). Desert falcons tend to be more number of pairs attempting to breed (cf. Postupalsky abundant in subtemperate and tropical regions than in 1974). Hence, the number of young fledged per success- temperate and Arctic regions (Table 3). Between 30°N ful pair was considered the most reliable criterion for and 30°S lanner, laggar and prairie are the most com- comparing breeding performance, although it probably mon large falcons within their respective ranges (Cade provided inflated estimates of absolute productivity 1982). Hence, there is reason to suspect interactions (Steenhof and Kochert 1982). between peregrines and congeners, with the latter being Globally, nesting habitat selectivity (in terms of a competitively dominant in tropical environments. This critical and easily measured variable, nest cliff height – is especially likely given that hierofalcons generally Jenkins 2000a) was negatively correlated with latitude breed earlier in the season than sympatric peregrines

Table 1. Mean breeding performance of peregrines across a range of latitudinal/climatic zones, based on data from recent studies of pristine or recovered populations (see Appendix for references for contributing studies).

Zone nn Young ﬂedged per studies pair-years Territorial pair Successful pair

High Arctic (65°N+) 6 849 1.81 2.54 Subarctic (55–64°N) 7 1494 1.57 2.40 Temperate (30–54°N/S) 19 2891 1.51 2.17 Subtropical (25–29°N/S)3 196 1.34 2.17 Tropical (B25°N/S) 3 90 1.42 1.92

360 ECOGRAPHY 24:3 (2001) Table 2. Mean nest site use, density and breeding performance of peregrine populations around the world (see Appendix for source data and references) in relation to latitude. Data provided are correlation coefﬁcients with sample sizes in parentheses (n.s.=not signiﬁcant, *pB0.05, **pB0.01, ***pB0.001).

Variable NorthernSouthern Old World New World Global hemisphere hemisphere

Mean nest cliff height – −0.82 (8)* ––−0.88 (10)** Minimum nest cliff height – −0.94 (9)*** –– –0.70 (12)* Inter-pair distance −0.79 (7)* −0.55 (6) n.s. −0.54 (9) n.s. – −0.40 (13) n.s. Clutch size 0.27 (7) n.s.– 0.69 (8) n.s. – 0.75 (10)* Young ﬂedged per territorial0.42 (24)*0.20 (11) n.s. −0.22 (17) n.s. 0.48 (15) n.s. 0.21 (35) n.s. pair Young ﬂedged per successful0.67 (26)*** 0.60 (11)* 0.05 (16) n.s. 0.77 (17)*** 0.57 (37)*** pair

(Cade 1960, Porter and White 1973, Jenkins 2000b) and considered minimum for non-competitive co-existence conceivably exclude them from potential nest sites (1.3 for linear measurements, 2.1 for non-linear mea- (Cade 1960, Porter and White 1973, Thomson 1984). surements – Hutchinson 1959, although see Simberloff and Boecklen 1981, Schoener 1984). Despite these structural similarities, diet overlap of sympatric peregrines and lanners in the Soutpansberg Evidence from South Africa was low (Morisita’sC=0.34 – Morisita 1959, Jenkins and Avery 1999), and the presence of close neighbour- Latitudinal trends in nesting habitat use, density and ing pairs of congeners did not significantly affect di- productivity of peregrines within South Africa generally etary niche width (Jenkins and Avery 1999). Hence, mirror global patterns. Certainly, peregrines are most there was no evidence of direct competition for food. common and widespread in the temperate southwest, However, although foraging habitat preferences of the and they are rare, patchy and restricted to high cliffs in two species (inferred from the habitat affinities of iden- the tropical northeast (Tarboton and Allan 1984, tified prey) were significantly different, they overlapped Jenkins 1994, 2000a). Lanners exhibit an opposite trend considerably (C=0.63 – Jenkins and Avery 1999). This (Jenkins 1994). Without experimental evidence, these overlap implies corresponding overlap in foraging contrasting distribution patterns cannot be conclusively ranges, and raises the possibility of interference compe- ascribed to either interspecific competition or resource tition between peregrines and lanners hunting in the limitation (Newton 1980, Connor and Simberloff 1986, same areas. Such interactions seem particularly likely Wiens 1989). However, sufficient observational data given that daily provisioning schedules of breeding have been accumulated from sympatric populations to pairs of the two species were not significantly different examine the relative strength of these hypotheses in (Jenkins 2000c), so foraging may have coincided tempo- light of theoretical predictions. rally as well as spatially.

Interspecific competition Phylogenetic relatedness, similarities in ecomorphology, significant niche overlap, correlated changes in niche dimensions with changing environmental conditions and high frequencies of interspecific aggression are features of sympatric populations that are considered to enhance the likelihood of interspecific competition, or to be circumstantial evidence of its effects (Wiens 1989). Morphologically, African peregrines and lanners are very similar (Jenkins 1995). Overall, peregrines have relatively smaller flight surfaces (so flight is more en- ergetically expensive) and larger feet (to facilitate the Fig. 1. Peregrine nesting habitat selectivity, measured in terms capture of aerial prey), hence they have a greater of mean and minimum nest cliff height, in relation to latitude. predilection than lanners to hunt from a perch, and to Data are from recent studies of pristine or recovered popula- target actively flying prey (Jenkins 1995). However, tions (see Appendix for source data and references). See Table comparisons of size and key food handling and flight 2 for correlation coefficients. The minimum cliff height data suggest a logarithmic relationship: y= –47.5 ln(x) +196.5 performance measurements of the two species (Table 4) (see curved line), which accounts for 71% of the variance in yield ratios consistently lower than values traditionally minimum cliff height.

ECOGRAPHY 24:3 (2001) 361 Table 3. The relative status of peregrines and similar, sympatric congeners (desert falcons) at different latitudes/climatic zones.

Latitude PeregrineRelative Desert falcon race status

Arctic tundrius Common/ Gyrfalcon uncommona peregrinus Common/ Lanner rareb Temperate anatum Uncommon/ Prairie commonc cassini Common/ absentd macropus Common/ absente minor Rare/ Lanner commonf Tropical peregrinator Rare/ Laggar commong

References: a Cade 1960, Court et al. 1988, Poole and Brom- ley 1988; b Cade et al. 1988, Massa et al. 1991, Manzi and Perna 1994; c Porter and White 1973, Cade et al. 1988; d McNutt et al. 1988; e Olsen and Olsen 1988b; f Steyn 1982, Brown et al. 1982, Jenkins 1994; g Ali and Ripley 1978, Cade 1982.

incidents. Of these 13 encounters, only eight involved intruding adults, and only one was recognized as confl- ict for territory between established pairs. In no instances did one species displace the other from a Fig. 2. Peregrine breeding performance, measured in terms of territory (but see Thomson 1982, Thomsett 1988), i.e. (a) mean clutch size and (b) the average number of young territory holders repelled intruders in all instances. In fledged per successful pair, in relation to latitude. Data are from recent studies of pristine or recovered populations (see the Soutpansberg, close neighbouring pairs of con- Appendix for source data and references). See Table 2 for geners did not significantly impair the breeding perfor- correlation coefficients. mance of peregrine or lanner pairs (Jenkins 2000b), further suggesting an absence of interference competi- Superficially, there was extensive overlap in nesting tion between the two species (cf. Korpima¨ki 1987, habitat use by the two falcon species in the Soutpans- Kostrzewa 1991). berg (Jenkins 2000a). However, typical nest sites were The accumulated evidence (Table 6) does not support significantly different in 10 of 32 measured variables, the contention that African peregrine populations are and over 90% of them could be correctly identified to limited by direct competition with sympatric lanners. species by discriminant function analysis (Jenkins Mean niche overlap (Holt 1987, Bosakowski et al. 2000a). The output of this analysis was used to measure 1992) between sympatric populations in the Soutpans- overlap in nesting habitat requirements (see Bosakowski et al. 1992), yielding a Morisita’s C value Table 4. Ratios (largest to smallest) comparing body mass of only 0.32. Thus, even given the fairly uniform cliff and measurements of key food handling and flight perfor- habitat available in the Soutpansberg (Jenkins 2000a), mance features of southern African peregrines and lanners microhabitat selection by nesting peregrine and lanner (calculated from means in Jenkins 1995). pairs was distinct. Males Females Territorial aggression between species is considered particularly good evidence of competition (Wiens Body mass 1.04 1.05 Bill length 1.01 1.02 1989). During this study, interactions between peregri- Wing span 1.10 1.09 nes and lanners comprised only 40 (5.3%) of 758 ag- Wing length 1.11 1.10 gressive incidents recorded in falcon territories, and Secondary length 1.21 1.19 Inner rectrix length 1.20 1.24 aggression between congeners was generally less fre- Tarsus length 1.06 1.02 quent than intraspecific aggression (Table 5). In condi- Toe 2 length 1.07 1.13 tions of direct sympatry (Soutpansberg, parts of the Toe 4 length 1.14 1.17 Wing loading 1.36 1.30 Orange River), interactions between peregrines and lan- Aspect ratio1.07 1.04 ners accounted for only 13 (5.4%) of 239 aggressive

362 ECOGRAPHY 24:3 (2001) Table 5. Incidence of inter- and intraspeciﬁc aggression recorded at peregrine and lanner nest sites in South Africa, with emphasis on the relative frequency of aggression between peregrines and intruding lanners and vice versa.

Population Frequencya of Frequencya of Aggression between congeners interspeciﬁc intraspeciﬁc aggressionb aggressionFrequency (n) % adults % pairs intruding intruding

Peregrines Cape Peninsula 0.11 0.040.01 (25) 4 0 Orange River 0.10 0.02 0.01 (6) 17 0 Soutpansberg 0.26 0.01 0.01 (4) 75 0 Lanners Orange River 0.20 0.02 0.04 (4) 100 25 Soutpansberg 0.20 0.030.01 (1) 0 0 a Mean number of incidents per hour of observation. b Involving species other than lanners/peregrines. berg was ca 0.4, well below the theoretical ‘‘competition ently in response to spring weather conditions. How- threshold’’ value of 0.6 (Bosakowski and Smith 1992, ever, large broods were fledged more frequently in this Bosakowski et al. 1992). However, the possibility that area, and the population was potentially more produc- existing distributions, behaviours and resource parti- tive in any given year (Jenkins 2000b). tioning are the result of past competitive interactions between the two species cannot be discounted (Wiens 1989). Evidence from other sympatric populations Relations between sympatric peregrine and desert falcon populations have been examined in Arctic (e.g. Resource limitation Cade 1960, Poole and Bromley 1988) and north temper- Food and nest sites are the resources that most com- monly limit raptor populations (Newton 1979). Re- Table 6. Factors limiting peregrines in tropical environments: duced tolerance of sub-optimal nesting habitats South African evidence for and against (a) competition with congeners and (b) resource deficiency. (Jenkins 2000a) effectively reduces nest site availability for tropical and subtropical peregrine populations in Evidence for Evidence against South Africa. However, given that nesting habitat qual- (a) proximate competition with sympatric lanners ity affects the foraging efficiency of resident pairs – 1. more common where 1. low diet overlap high nest cliffs contribute to foraging success by provid- lanners are rare 2. diet unaffected by ing perch-hunting falcons with an effective height ad- 2. similar ecomorphology close neighbouring congeners vantage over their prey (Jenkins 2000d) – nest site 3. moderate overlap in foraging 3. low nesting habitat selectivity may be expected to correlate negatively with habitats overlap and times food abundance or availability. In this way, popula- 4. breeding unaffected tions proximately limited by nest site availability may by close neighbouring be ultimately limited by food and its defining effect on congeners the quality of breeding areas (Jenkins 1991). 5. little aggression General correlations between falcon productivity and between congeners indices of environmental productivity emphasize the (b) resource deficiency in the Soutpansberg (vs the Cape importance of food in limiting peregrine populations in peninsula) 1. more stringent nesting 1. lower egg and South Africa (Jenkins 1991) and other areas (e.g. Thiol- habitat requirements hatchling mortality lay 1988, Ratcliffe 1993). Circumstantial evidence of 2. lower density of breeding 2. higher mean annual relative food shortage at tropical (Soutpansberg) nests pairs productivity 3. lower hunting rates in the in relation to temperate (Cape peninsula) nests included vicinity of nest sites lower provisioning rates, inferior parental care and the 4. lower provisioning rates at suggestion of greater foraging effort by Soutpansberg nests 5. less parental attendance at pairs (Jenkins 2000c). Nestling growth rates were nests slightly lower in the Soutpansberg, although no physi- 6. greater foraging effort by cal signs of food deprivation were observed in either breeding pairs area. Productivity of the two populations was not 7. inferior nestling growth rates 8. lower frequency of large significantly different, and was subject to considerable broods of young inter-annual fluctuation on the Cape Peninsula, appar-

ECOGRAPHY 24:3 (2001) 363 ate environments (e.g. Porter and White 1973, Massa et more specialized bird predator, its distribution should al. 1991, Manzi and Perna 1994). These studies gener- be more directly influenced by the dynamics of prey ally found dietary segregation between falcon popula- populations, and more sensitive to the effects of habitat tions. However, in two cases at least, the species with structure (particularly nest cliff height) on the vulnera- the most specialized nest site requirements (Alaskan bility or accessibility of aerial prey. gyrfalcons and peregrines in northern Italy) were con- Bird breeding seasons are more synchronous and sidered to limit populations of their less selective coun- more productive with increasing latitude (Ricklefs 1980, terparts (peregrines and lanners respectively) by Wyndham 1986). This concentration of the breeding restricting their access to otherwise usable habitat effort is likely to provide flushes of displaying, provi- (Cade 1960, Manzi and Perna 1994). This interpretation sioning and newly-fledged birds, particularly vulnerable suggests that African peregrines may competitively ex- to predation by peregrines. Exploitation of such an clude lanners from certain areas, rather than the abundant source of available prey may be what permits reverse. peregrines in Arctic environments to sustain higher Overall, the results of this study support the resource breeding densities and achieve greater breeding success deficiency hypothesis, with food limitation presenting than populations elsewhere (Table 1, Fig. 1), even the most convincing explanation for the rarity of pere- without the benefit of high cliffs to facilitate foraging grines in tropical environments (Table 6). (Fig. 2, Jenkins 2000d). Conversely, increasingly asea- sonal and less productive avian breeding towards the tropics (Ricklefs 1980, Wyndham 1986), may lower the overall quality of the peregrine’s prey base (in terms of Latitude, avian productivity and prey its general vulnerability to attack), restricting breeding availability pairs to optimal habitats (high cliffs) which are limiting in the environment. Bird-eating raptors generally have low hunting success This hypothesis predicts that prey breeding regimes and may experience low energy returns per unit forag- are particularly influential in the timing and success of ing time (Temeles 1985). Success in strikes at flying peregrine breeding cycles at high latitudes. Recent stud- birds is strongly influenced by qualities of the predator, ies of Arctic peregrines suggest that this is the case. its prospective prey and the structure of the underlying Falcon breeding is timed to coincide with the fledging habitat (e.g. Kenward 1978, Jenkins 2000d). Such sensi- period of local passerine populations (Falk et al. 1986, tivity accentuates the distinction between prey abun- Court et al. 1988), and fledgling or juvenile birds com- dance and prey availability (in terms of its vulnerability prise ‘‘the bulk’’ (perhaps 60–70%) of prey delivered to to attack – Temeles 1985). This discrepancy may be active nests in some areas (Court et al. 1988, Rosenfield further exaggerated in the case of large falcons, which et al. 1995). typically hunt birds at higher speeds than other bird- The ‘‘prey productivity’’ theory is largely supported eating raptors (e.g. Accipiter hawks), in more open, (but not tested) by data from South Africa. Avian three-dimensional airspace (Cade 1982, Jenkins 1995, breeding seasons are generally shorter, more synchro- 2000d). Obvious examples of the dependence of breed- nized and more productive in the temperate southwest- ing falcons on particularly abundant and vulnerable ern parts of the country than in the subtropical/tropical sources of prey are the Eleanora’s F. eleanorae and northeast (Winterbottom 1963, Harrison et al. 1997). sooty F. concolor populations of the Mediterranean and Correspondingly, peregrines preyed on juvenile birds Red Seas. The siting, timing and success of breeding by more frequently on the Cape peninsula than in the these specialized, colonial falcons are highly sensitive to Soutpansberg (Jenkins 1998), and the proportion of variations in the route, timing and extent of small bird juvenile birds taken by breeding pairs on the Cape migration on the Eurasia – Africa flyway (Walter 1979, peninsula was sensitive to correlates of prey breeding Gaucher et al. 1995). seasonality and success, as was the overall productivity Peregrines often occupy a broader food niche than of the population (Jenkins and Avery 1999, Jenkins sympatric congeners (Jenkins and Avery 1999), but the 2000b). latter consistently take a higher proportion of largely While other studies have noted that tropical popula- terrestrial or non-avian prey (e.g. Cade 1960, Porter tions of widespread raptor taxa appear to be limited by and White 1973, Massa et al. 1991, Jenkins and Avery food (e.g. Simmons 1986, Beissinger 1990), this ‘‘prey 1999). This may be explained in terms of subtle differ- productivity’’ hypothesis is unique in linking predator ences in flight morphology and energetics (e.g. Jenkins density to prey availability via a changeable definition 1995). Such differences promote a relatively sedentary of habitat quality. Ultimately, peregrines are limited by hunting mode in peregrines (and a tendency to take specialized morphology and behavioural stereotypy active, aerial prey) and more active foraging by desert which restrict foraging mode, and place increasing em- falcons (and a greater capacity to take less mobile prey phasis on habitat quality (in terms of physical charac- on the ground) (Jenkins 1995). As the peregrine is a teristics that favour aerial hunting) as food conditions

364 ECOGRAPHY 24:3 (2001) (in terms of the abundance and/or vulnerability of prey) References decline. Obviously, nest cliff height and the effect of Ali, S. and Ripley, S. D. 1978. Handbook of the birds of India latitude on breeding seasonality are not the only deter- and Pakistan. Vol. 1. – Oxford Univ. Press. minants of prey availability for peregrines. Areas where Ambrose, R. E. et al. 1988. Changes in the status of peregrine birds of a suitable size are otherwise particularly abun- falcon populations in Alaska. – In: Cade, T. J. et al. (eds), dant, concentrated or vulnerable to aerial attack may Peregrine falcon populations: their management and recovery. The Peregrine Fund, Idaho, pp. 73–82. support fairly dense and productive peregrine popula- Baker, A. J. et al. 2000. The orange-breasted falcon Falco tions, irrespective of topography and latitude. However, deiroleucus in Mesoamerica: a vulnerable, disjunct popula- this study has shown that, on a broad scale, these two tion? – Bird Conserv. Int. 10: 29–40. Beebe, F. L. 1974. Field studies of the Falconiformes of the parameters clearly explain the pattern of peregrine dis- British Columbia: vultures, hawks, falcons, eagles. – British tribution worldwide. Columbia Prov. Mus. Occas. Pap. 17: 1–163. Few avian predators are as widespread or as special- Beissinger, S. R. 1990. Experimental brood manipulations and ized in their manner of securing prey as the peregrine. the monoparental threshold in snail kites. – Am. Nat. 136: 20–38. Consequently, few species are likely to exhibit the same Bird, D. M. and Weaver, J. D. 1988. Peregrine falcon popula- degree of responsiveness to global patterns of produc- tions in Ungava Bay, Quebec, 1980–1985. – In: Cade, T. tivity. However, parallels in peregrine and osprey Pan- J. et al. (eds), Peregrine falcon populations: their management and recovery. The Peregrine Fund, Idaho, pp. 45–49. dion haliaetus biology suggest that the ideas presented Bosakowski, T. and Smith, D. G. 1992. Comparative diets of here may be more widely applicable. The osprey has a sympatric nesting raptors in the eastern deciduous forest cosmopolitan distribution and feeds entirely on fish, biome. – Can. J. Zool. 70: 984–992. caught in spectacular, vertical dives (del Hoyo et al. Bosakowski, T., Smith, D. G. and Speiser, R. 1992. Niche overlap of two sympatric-nesting hawks Accipiter spp. in the 1994). Foraging efficiency is profoundly influenced by New Jersey–New York Highlands. – Ecography 15: 358– the physical structure of water bodies and the abun- 372. dance and dispersion of prey (Swenson 1979, Poole Bradley, M. et al. 1997. Influence of weather on breeding success of peregrine falcons in the arctic. – Auk 114: 786–791. 1989), and clutch sizes and provisioning rates at nests Bromley, R. G. and Mathews, S. B. 1988. Status of the peregrine are positively correlated with latitude (Poole 1982, falcon in the Mackenzie River Valley, Northwest Territories 1989). 1969–1985. – In: Cade, T. J. et al. (eds), Peregrine falcon populations: their management and recovery. The Peregrine Fund, Idaho, pp. 59–63. Brown, C. J. and Cooper, T. G. 1987. The status of cliff-nesting raptors on the Waterberg, SWA/Namibia. – Madoqua 15: 243–249. Jack-of-all-trades or master of one? Brown, L. H., Urban, M. K. and Newman, K. 1982. The birds Specialization and the distribution of large of Africa. – Academic Press. Cade, T. J. 1960. Ecology of the peregrine and gyrfalcon falcons populations in Alaska. – Univ. California Publ. Zool. 63: 151–290. Functionally, peregrines and desert falcons represent Cade, T. J. 1982. The falcons of the world. – Collins. variations on a common theme, and are similarly de- Cade, T. J. et al. (eds) 1988. Peregrine falcon populations: their signed as fast-flying predators of birds. These qualities management and recovery. – The Peregrine Fund, Idaho. are emphasized in the peregrine group to the extent that Connor, E. F. and Simberloff, D. 1986. Competition, scientific method, and null models in ecology. – Am. Sci. 74: 155–162. its success in any environment is dependent on the Court, G. S., Gates, C. C. and Boag, D. A. 1988. Natural history concentration or vulnerability of the avian prey base, of the peregrine falcon in the Keewatin District of the and the effects of topography on foraging conditions Northwest Territories. – Arctic 41: 17–30. Crick, H. Q. P. and Ratcliffe, D. A. 1995. The peregrine Falco (Beebe 1974). The desert falcons have retained a degree peregrinus breeding population of the United Kingdom in of plasticity in foraging mode and local nesting habitat 1991. – Bird Study 42: 1–19. requirements. The two groups show opposing responses Czechura, G. V. and Debus, S. J. S. 1985. The black falcon Falco to postulated latitudinal trends in prey availability: subniger: a summary of information and comparison with the brown falcon Falco berigora. – Aust. Bird Watcher 11: niche contraction and rarity (peregrines) vs niche ex- 80–91. pansion and relative abundance (hierofalcons) with de- del Hoyo, J., Elliot, A. and Sargatal, J. (eds) 1994. Handbook creasing resource availability towards the Equator. of the birds of the world. Vol. 2. New World vultures to guineafowl. – Lynx Edicions, Barcelona. Ellis, D. H. 1988. Distribution, productivity, and status of the Acknowledgements – Thanks to Dave Allan, Zelda Bate, Alan peregrine falcon in Arizona. – In: Cade, T. J. et al. (eds), Stephenson, Tim Wagner and Anthony van Zyl for their help Peregrine falcon populations: their management and recov- in the field. The ideas presented here benefited greatly from ery. The Peregrine Fund, Idaho, pp. 87–94. discussion or correspondence with Dave Allan, Chris Brown, Enderson, J. H., Craig, G. R. and Burnham, W. A. 1988. Status Rob Davies, Alan Kemp, Gerard Malan, Roy Siegfried, Rob of peregrines in the Rocky Mountains and Colorado Simmons, Alan Stephenson, Tim Wagner and Anthony van Plateau. – In: Cade, T. J. et al. (eds), Peregrine falcon Zyl. Tom Cade, Morne´ du Plessis and Antoni Milewski made populations: their management and recovery. The Peregrine useful comments on earlier drafts of the manuscript. This Fund, Idaho, pp. 83–86. project was partly funded by the Raptor Conservation Group, Falk, K. and Mo¨ller, S. 1988. Status of the peregrine falcon in Steve Phelps of Peregrine Properties, and the Foundation for south Greenland: population density and reproduction. – Research Development. In: Cade, T. J. et al. (eds), Peregrine falcon populations: their

ECOGRAPHY 24:3 (2001) 365 management and recovery. The Peregrine Fund, Idaho, pp. Peregrine falcon populations: their management and recov- 37–43. ery. The Peregrine Fund, Idaho, pp. 159–172. Falk, K., Mo¨ller, S. and Burnham, W. A. 1986. The peregrine Manzi, A. and Perna, P. 1994. Relationships between peregrine falcon Falco peregrinus in south Greenland: nesting require- and lanner in the Marches (central Italy). – In: Meyburg, ments, phenology and prey selection. – Dansk. Orn. Foren. B.-U. and Chancellor, R. D. (eds), Raptor conservation Tidsskr. 80: 113–120. today. WWGBP, Berlin, pp. 157–162. Fasce, P. and Fasce, L. 1988. Peregrines in northern Italy: Massa, B. et al. 1991. Il lanario falco biarmicus feldeggi numbers, breeding, and population dynamics. – In: Cade, (Schlegel) in Italia: status, biologia e tassonomia. – Natural- T. J. et al. (eds), Peregrine falcon populations: their manage- ista Sicil. S. IV, XV: 27–63. ment and recovery. The Peregrine Fund, Idaho, pp. 711– Mattox, W. G. and Seegar, W. S. 1988. The Greenland peregrine 718. falcon survey, 1972–1985, with emphasis on recent popula- Gaucher, P., Thiollay, J. M. and Eichaker, X. 1995. The sooty tion status. – In: Cade, T. J. et al. (eds), Peregrine falcon falcon Falco concolor on the Red Sea coast of Saudi Arabia: populations: their management and recovery. The Peregrine distribution, numbers and conservation. – Ibis 137: 29–34. Fund, Idaho, pp. 27–36. Harrison, J. A. et al. (eds) 1997. The atlas of southern African McNutt, J. W. et al. 1988. Distribution and status of the birds. Vol. 2. Non-passerines. – Birdlife South Africa, peregrine falcon in South America. – In: Cade, T. J. et al. Johannesburg. (eds), Peregrine falcon populations: their management and Hartley, R. 1992. Status and productivity of peregrines in low recovery. The Peregrine Fund, Idaho, pp. 237–249. latitudes. – Gabar 7: 38–40. Mendelsohn, J. M. 1988. The status and biology of the peregrine Hartley, R. et al. 1993. Notes on the breeding biology, hunting in the Afrotropical region. – In: Cade, T. J. et al. (eds), behaviour, and ecology of the Taita falcon in Zimbabwe. – Peregrine falcon populations: their management and recov- J. Raptor Res. 27: 133–142. ery. The Peregrine Fund, Idaho, pp. 297–306. Hartley, R., Newton, I. and Robertson, M. 1995. Organochlo- Morisita, M. 1959. Measuring of interspecific association and rine residues and eggshell thinning in the peregrine falcon similarity between communities. – Mem. Fac. Sci. Kyushu Falco peregrinus minor in Zimbabwe. – Ostrich 66: 69–73. Univ. Ser. E (Biol.) 3: 65–80. Heredia, B. et al. 1988. Status, ecology, and conservation of the Nelson, R. W. 1990. Status of the peregrine falcon Falco peregrine falcon in Spain. – In: Cade, T. J. et al. (eds), peregrinus pealei, on Langara Island, Queen Charlotte Is- Peregrine falcon populations: their management and recov- lands, British Columbia, 1968–1989. – Can. Field Nat. 104: ery. The Peregrine Fund, Idaho, pp. 219–226. 193–199. Holt, R. D. 1987. On the relation between niche overlap and Newton, I. 1979. Population ecology of raptors. – T. and A. competition: the effect of incommensurable niche dimen- D. Poyser. sions. – Oikos 48: 110–114. Newton, I. 1980. The role of food in limiting bird numbers. – Hunt, W. G. et al. 1988. Nesting peregrines in Texas and Ardea 68: 11–30. northern Mexico. – In: Cade, T. J. et al. (eds), Peregrine Olsen, P. D. and Olsen, J. 1988a. Breeding of the peregrine falcon populations: their management and recovery. The falcon Falco peregrinus: I. Weather, nest spacing and territory occupancy. – Emu 88: 195–201. Peregrine Fund, Idaho, pp. 115–121. Olsen, P. D. and Olsen, J. 1988b. Population trends, distribu- Hutchinson, G. E. 1959. Homage to Santa Rosalia, or why are tion, and status of the peregrine falcon in Australia. – In: there so many kinds of animals? – Am. Nat. 93: 145–159. Cade, T. J. et al. (eds), Peregrine falcon populations: their Jenkins, A. R. 1991. Latitudinal prey productivity and potential management and recovery. The Peregrine Fund, Idaho, pp. density in the peregrine falcon. – Gabar 6: 20–24. 255–274. Jenkins, A. R. 1994. The influence of habitat on the distribution Olsen, P. D. and Olsen, J. 1989a. Breeding of the peregrine and abundance of peregrine and lanner falcons in South falcon Falco peregrinus: II. Weather, nest quality and the Africa. – Ostrich 65: 281–290. timing of egg laying. – Emu 89: 1–5. Jenkins, A. R. 1995. Morphometrics and flight performance of Olsen, P. D. and Olsen, J. 1989b. Breeding of the peregrine southern African peregrine and lanner falcons. – J. Avian falcon Falco peregrinus: III. Weather, nest quality and Biol. 26: 49–58. breeding success. – Emu 89: 6–14. Jenkins, A. R. 1998. Behavioural ecology of peregrine and Pepler, D., van Hensbergen, H. J. and Martin, R. 1991. Breeding lanner falcons in South Africa. – Unpubl. Ph.D. thesis, density and nest site characteristics of the peregrine falcon Univ. Cape Town, Cape Town. Falco peregrinus minor in the southwestern Cape, South Jenkins, A. R. 2000a. Characteristics of peregrine and lanner Africa. – Ostrich 62: 23–28. falcon nesting habitats in South Africa. – Ostrich, in press. Poole, A. F. 1982. Brood reduction in temperate and sub-trop- Jenkins, A. R. 2000b. Factors affecting breeding success of ical Ospreys. – Oecologia 53: 111–119. peregrine and lanner falcons in South Africa. – Ostrich, in Poole, A. F. 1989. Ospreys: a natural and unnatural history. – press. Cambridge Univ. Press. Jenkins, A. R. 2000c. Variation in the quality of parental care Poole, K. G. and Bromley, R. G. 1988. Interrelationships within at falcon nests in South Africa as evidence for postulated a raptor guild in the central Canadian Arctic. – Can. J. Zool. differences in food availability. – Ardea 88: 17–32. 66: 2275–2282. Jenkins, A. R. 2000d. Hunting mode and success of African Porter, R. D. and White, C. M. 1973. The peregrine falcon in peregrines Falco peregrinus minor: does nesting habitat Utah: emphasizing ecology and competition with the prairie quality affect foraging efficiency? – Ibis 142: 235–246. falcon. – Brigham Young Univ. Sci. Bull. 18: 1–74. Jenkins, A. R. and Avery, G. M. 1999. Diets of breeding Porter, R. D. et al. 1988. Status and reproductive performance peregrine and lanner falcons in South Africa. – J. Raptor of marine peregrines in Baja California and the Gulf of Res. 33: 190–206. California, Mexico. – In: Cade, T. J. et al. (eds), Peregrine Kenward, R. E. 1978. Hawks and doves: factors affecting falcon populations: their management and recovery. The success and selection in goshawk attacks on woodpigeons. Peregrine Fund, Idaho, pp. 105–114. – J. Anim. Ecol. 47: 449–460. Postupalsky, S. 1974. Raptor reproductive success: some prob- Korpima¨ki, E. 1987. Dietary shifts, niche relationships and lems with methods, criteria and terminology. – Raptor Res. reproductive output of coexisting kestrels and long-eared Rep. 2: 21–31. owls. – Oecologia 74: 277–285. Pruett-Jones, S. G., White, C. M. and Devine, W. R. 1981. Kostrzewa, A. 1991. Interspecific interference competition in Breeding of the peregrine falcon in Victoria. – Emu 80: three European raptor species. – Ethol. Ecol. Evol. 3: 253–269. 127–143. Ratcliffe, D. A. 1993. The peregrine falcon. – T. & A. D. Poyser. Lindberg, P., Schei, P. J. and Wikman, M. 1988. The peregrine Ricklefs, R. E. 1980. Geographic variation in clutch size among falcon in Fennoscandia. – In: Cade, T. J. et al. (eds), passerine birds: Ashmole’s hypothesis. – Auk 97: 38–49.

366 ECOGRAPHY 24:3 (2001) Rosenﬁeld, R. N. et al. 1995. Prey of peregrine falcons breeding Thiollay, J. M. 1988. Prey availability limiting an island in west Greenland. – Condor 97: 763–770. population of peregrine falcons in Tunisia. – In: Cade, T. Schoener, T. W. 1984. Size differences among sympatric, bird- J. et al. (eds), Peregrine falcon populations: their manage- eating hawks: a world wide survey. – In: Strong, D. R. et ment and recovery. The Peregrine Fund, Idaho, pp. 701– al. (eds), Ecological communities, conceptual issues and the 710. evidence. Princeton Univ. Press, pp. 254–281. Thomsett, S. 1988. Distribution and status of the peregrine in Simberloff, D. and Boecklen, W. 1981. Santa Rosalia reconsid- Kenya. – In: Cade, T. J. et al. (eds), Peregrine falcon ered: size ratios and competition. – Evolution 35: 1206– populations: their management and recovery. The Peregrine 1228. Fund, Idaho, pp. 289–295. Simmons, R. 1986. Food provisioning, nestling growth and Thomson, W. R. 1982. Peregrines in Zimbabwe. – Honeyguide experimental manipulation of brood size in the African 111/112: 26–28. redbreasted sparrowhawk Accipiter ruﬁ6entris. – Ornis Thomson, W. R. 1984. Comparative notes on the ecology of Scand. 17: 31–40. peregrine, lanner and Taita falcons in Zimbabwe. – In: Steenhof, K. and Kochert, M. N. 1982. An evaluation of Mendelsohn, J. M. and Sapsford, C. W. (eds), Proceedings methods used to estimate raptor nesting success. – J. Wildl. of the 2nd symposium on African predatory birds. Natal Manage. 46: 885–893. Bird Club, Durban, pp. 15–18. Steyn, P. 1982. Birds of prey of southern Africa. – David Philip, Walter, H. 1979. Eleanora’s falcon: adaptations to prey and Cape Town. habitat in a social raptor. – Univ. of Chicago Press. Swenson, J. E. 1979. The relationship between prey species Wiens, J. A. 1989. The ecology of bird communities. Vol. 2. – ecology and dive success in ospreys. – Auk 96: 408–412. Cambridge Univ. Press. Tarboton, W. R. and Allan, D. G. 1984. The status and Winterbottom, J. M. 1963. Avian breeding seasons in southern conservation of birds of prey in the Transvaal. – Transvaal Africa. – In: Sibley, C. G. (ed.), Proc. of the XIII interna- Mus. Monogr. 3: 1–115. tional ornithological congress. Vol. 2. American Ornitholo- Temeles, E. J. 1985. Sexual size dimorphism of bird-eating gist’s Union, Louisiana, pp. 640–648. hawks: the effect of prey vulnerability. – Am. Nat. 125: Wyndham, E. 1986. Length of birds’ breeding seasons. – Am. 485–499. Nat. 128: 155–164.

Appendix. Key data describing nest site selectivity and breeding success of peregrine populations around the world. Numbered variables are: 1 – minimum/mean nest cliff height (m), 2 – mean inter-pair distance (km), 3 – mean clutch size, 4 – mean number of young ﬂedged per territorial pair, 5 – mean number of young ﬂedged per successful pair.

Latitude 1 2 3 4 5 Reference

969°N – –– 1.86 2.36 Bromley and Mathews 1988 969°N – – – 1.63 2.73 Ambrose et al. 1988 968°N ––3.35? 1.622.47 Lindberg et al. 1988 967°N 27/––– 2.40 2.96 Mattox and Seegar 1988 965°N ––– 2.15 2.67 Ambrose et al. 1988 965°N ––– 1.22 2.03 Lindberg et al. 1988 962°N7/9203.3 3.64 2.03 2.76 Court et al. 1988 961°N ––– 1.76 2.72 Falk and Mo¨ller 1988 959°N – –– 2.56 2.69 Bird and Weaver 1988 958°N B20/940– 3.46 0.98 2.08? Ratcliffe 1993 957°N B20/945 – – 1.35 2.15 Ratcliffe 1993 956°N B20/9355.4? – 1.23 2.15 Ratcliffe 1993 955°N B11/920 94.2 3.42 1.09 2.27 Ratcliffe 1993 954°N – 94.0 3.45 1.41 2.43 Ratcliffe 1993 954°N – – 3.24 1.42 2.41 Crick and Ratcliffe 1995 954°N ––––2.24 Nelson 1990 953°N B11/935 94.5 3.40 1.22 2.05 Ratcliffe 1993 951°N – 94.1 – 1.51 2.29 Ratcliffe 1993 945°N – 11.6 – 1.99 2.50 Fasce and Fasce 1988 944°N ––– 1.20 2.10 Fasce and Fasce 1988 940°N ––– 1.73 – Heredia et al. 1988 936°N – –– 1.73 2.14 Enderson et al. 1988 936°N – – – – 1.74 Ellis 1988 932°N – –– – 1.36 Ellis 1988 932°N ––– 1.01 1.99 Hunt et al. 1988 928°N – –– 1.08 2.14 Porter et al. 1988 926°N – – – 1.25 1.95 Hunt et al. 1988 910°S 100/115? –– 1.44 1.50 McNutt et al. 1988 918°S ––– 1.45 1.83 Hartley et al. 1995 920°S50/85 5.1 –––Brown and Cooper 1987 923°S75/114 9.7 3.14 1.36 2.44 Jenkins 2000a, b 928°S25/826.3 – 1.70 2.42 Jenkins 2000a, b 930°S ––– 2.72 2.75 McNutt et al. 1988 934°S 20/79 5.6 2.78 1.11 2.29 Jenkins 2000a, b 936°S20/– 4.7? – 2.00 2.20 Olsen and Olsen 1988b 936°S 10/25 4.3 – 1.44 2.16 Olsen and Olsen 1988a, 1989a, b 937°S 4/29 2.81 1.21 2.00 Pruett-Jones et al. 1981 942°S7/49 – – 1.90 2.50 Olsen and Olsen 1988b 948°S – –– 1.79 2.47 McNutt et al. 1988

ECOGRAPHY 24:3 (2001) 367