Sex and plumage-type ratios of the Lesser Magellan Goose in southern Chile

Transcription

1 Sex and plumage-type ratios of the Lesser Magellan Goose in southern Chile W.R. S IE G F R IE D. P.A.R. H O C K E Y. P.G. R Y A N and A.L. B O SM A N Introduction Four of the five species of South A m erican sheldgeese (genus C hloëphaga) are terrestrial grass-eaters; the exception being the Kelp G oose C. hybrida (D elacour 1954). T hree of the four terrestrial species breed in southern P atagonia, including T ierra del Fuego (Johnson 1965; H um phrey et al. 1970). T hey are the A shy-headed G oose C. poliocephala, R u d d y -h ead ed G oose C. rubidiceps and M agellan G oose C. picta which has two races, the L esser M agellan G oose C. p. picta on the South A m erican m ainland, and the G reater M agellan G oose C. p. leucoptera on the Falkland Islands. T he M agellan G oose exhibits m arked sexual plum age dim orphism, but in the o th er two species the sexes are sim ilar. The L esser M agellan G oose also differs from the o ther species in that adult m ales present a range o f colour phases, from individuals with com pletely w hite underparts to birds heavily barred o r blotched w ith black and w hite. In the G re a te r M agellan G oose, by contrast, the undcrparts of adult m ales invariably are pure w hite. Fem ales in general show little variation, but the colour of the head o f the L esser M agellan G oose ranges betw een uniform reddish and dull grey cinnam on (Johnson 1965; B lake 1977). It has been know n for a long tim e that the extrem e colour phases presented by the Lesser M agellan G oose are m ore-or-less segregated in different regions and at different seasons; the barred types being m ost abundant in the south and th e w hite type in the north (Johnson 1965; H um phrey et al. 1970). H ow ever, quantitative data supporting this are scarce. T his pap er sum m arises the results of counts of L esser M agellan, A shy-headed and R uddy-headed G eese m ade in the course o f m otoring over som e 800 km of roads in southern C hile, including T ierra del Fuego, during N ovem ber T hese roadside counts gave quantitative data on relative abundance, group size and inter-specific asso ciatio n fo r all th re e species, and sex and plum age-type ratios for the L esser M agellan G oose betw een ca 51 S and 54 S, close to the southern limit of their breeding distribution. Methods Sheldgeese w ere counted w henever seen w ithin approxim ately 500 m o f the roadside. H ence the relative abundance of sheldgeese is expressed as num bers of birds per km 3. T he m ain routes traversed extended from T orres del P aine, in the north, via P uerto N atales and Punta A renas, to Fort B ulnes in the south; from P unta A renas, across the Brunsw ick P eninsula, to O tw ay Sound; from P unta A renas, along the Straits of M agellan, north-east to Punta D elgada; and, on T ierra del Fuego, from Porvenir north-east to P unta E spora (see Figure 1). T he shelcfgeese on the route betw een T o r res del Paine and P unta A renas w ere surveyed only once, but the southern routes w ere travelled m ore frequently. In such cases, how ever, the surveys w ere restricted to separate days. A total o f 310 encounters w ith sheldgeese w ere recorded in the extended road-strip, covering 808 km. S heldgeese w ere o b serv ed carefully, using binoculars and telescopes, and their behaviour, in term s of feeding, nesting and association betw een sexes, was recorded. T he birds' habitats and any association betw een the birds and dom estic livestock w ere also recorded. T h ree categories of m ales w ere recognised based on the extent of barring on their underparts: com pletely w h ite; in te rm e d ia te ; a n d. co m p le te ly barred. D elacour (1954) im plies th at belly colour o f im m atures reflects the adult colour phase, and thus the presence of im m ature and sub-adult birds in flocks should not bias analyses of plum age-type ratios. Fem ales w ere also grouped in three cate g o ries, based on h ead colo ratio n : greyish cinnam on; interm ediate; and, reddish cinnam on. B irds w ere classified as 'in d eterm in ate' w hen p o o r light, cover or som e o th er factor prevented them from being categorised accurately. B irds w ere adjudged to be paired w hen individual m ales and fem ales associated and m oved closely together, usually within a m etre of each other. In som e flocks it was not possible to determ ine paired/unpaired ratios: such flocks w ere used only in analyses of sex ratio and group size. 15 W ildfowl 34 (I4SX ): 15-21

2 16 W.R. Siegfried, P.A.R. H ockey, P.G. R yan a n d A.L. Bosnian Fig. 1. Map of the study area in southern Chile. Dotted lines indicate routes taken during surveys. Results Relative abundance T otals o f 1,941 (95% ) L esser M agellan, 78 (4% ) A shy-headed and 20 (1% ) R uddyheaded G eese w ere recorded. T he L esser M agellan G oose was m arkedly m ore abund an t in the n orth o f th e survey area (T orres del Paine to P u erto N atales) and in n o rth ern T ie rra del F uego th an elsew here. R uddy-headed and A shy-headed G eese w ere uncom m on in all th ree areas (T able 1). H ighest densities of th e A shy-headed G oose w ere found on the B runsw ick P eninsula, norm ally in association with indigenous forest patches. Several pairs w ere found breeding along the interm ittently forested R io San Juan to the south of Fort B ulnes. T he R uddy-headed G oose was m ost ab u n d an t, although still rare, in the grasslands o f north ern T ierra del Fuego, w here th e A shy-headed G oose was absent. B ased on relative abundance o f the geese Table 1. Relative abundance of three species of sheldgeese in southern Patagonia, November Survey route A rea surveyed (km2) Lesser Magellan Goose Mean density of birds per krrr Ashy-headed Goose Ruddy-headed Goose Torres del Paine - Puerto Natales < 0.01 Puerto Natales - Punta Arenas P. A renas A irport - Punta Delgada Brunswick Peninsula Porvenir - Punta Espora

3 Lesser Magellan geese in Chile 17 Table 2. Percentage frequency of group size and the proportions (percentages) of paired birds in the Lesser Magellan Goose in three regions of southern Patagonia, November G roup size Sample Proport ion of paired birds Num ber of pairs observed size with broods of small >10 (no. groups) Males Females Overall young North Central 60 IS South 71 IS species, th ree regions w ere recognised a posteriori within o u r study area: N orth (T orres del Paine - P uerto N atales); C entral (m ainland south of P u erto N atales); and. South (T ierra del Fuego) (Figure 1). T hese regions w ere used in subsequent analyses. G roup size G roup sizes of A shy-headed and R uddyheaded G eese varied betw een one and 24 and one and eight birds, respectively. D ata bases for these tw o species are too small to allow fu rth er statistical analysis aim ed at revealing any geographical differences in group size. In the L esser M agellan G oose, group size ranged betw een tw o and 152 birds. In all regions, a group size of tw o was m ost frequent (T able 2) and in only one instance did these observations not relate to p resum ed m ated pairs. T here was a tendency for groups to be largest in the C entral region w here density was lowest (T ables 1 and 2), corresponding w ith a low er p roportion of paired birds in this region. H ow ever, groupsize frequency was not significantly different betw een the regions (X 2= 6.05, d f= 4, P > ). Sex ratio and plum age types Sex ratios o f the L esser M agellan G oose w ere very close to parity in all th ree regions: N o rth, m ales = 52.2% ; C entral, m ales = 47.7% ; and, South, m ales = 52.5%. T here was a clear latitudinal cline in male plum age-type frequency, w ith w hite birds predom inating in the N orth and barredbellied birds in the South. Interm ediate type birds w ere m ore frequent in the N orth and C entral regions than in the South (X 2= 29.28, d f= 2, PC 0.002) (T able 3). T he p attern am ong fem ales was less obvious. T here was a clear gradient of grey-headed birds, d e c re a s in g to w a rd s h ig h e r la titu d e s, although interm ediate fem ales w ere num erically dom inant in all regions and becam e increasingly so in the South (T able 3). The proportion of paired red-headed fem ales in the C entral region was g reater than expected (X 2= 10.34, d f= 2, PcO.O l). These trends are reflected in the frequency distributions of the plum age-type com binations of paired birds in the th ree regions (T able 4). A ssum ing that solitary m ales indicated actively breeding birds, th ere was no difference in their abundance relative to the occurrence o f m ated, b u t not actively breeding, m ales in each of th e th ree regions (X 2 = 2.86, d f= 2, P > ). Sim ilarly, there was no difference betw een plum age-type ratios of breeding m ales and m ated m ales in th e S outh region (X 2= 2.9 2, d f= 2, P > ), or betw een sim ilar ratios of w hite and interm ediate m ales in th e N orth region (X 2= 0.06, d f = l, P > 0.05). Sam ple sizes for barred m ales in the n o rth ern region, and for all plum age-types in th e C entral region w ere too small for statistical analysis. Table 3. Percentage frequency of three plumage types in male and female populations of the Lesser Magellan Goose in three regions of southern Patagonia, November W hite Males Plumage-types Interm ediate Barred N Greyish Females Plumage-types Interm ediate Reddish N North Central South

4 18 W.R. Siegfried, P.A.R. H ockey, P.G. Ryan an d A.L. Bosnian Table 4. Percentage frequency of the nine possible plumage-type pairings of the Lesser Magellan Goose in three regions of southern Patagonia, November Male plumage-type White Interm ediate Barred Total North female Greyish plumage Interm ediate type Reddish Total n=58 pairs Central female Greyish plumage Interm ediate type Reddish Total n=58 pairs South female Greyish plumage Interm ediate type Reddish Total n=151 pairs A ssociations between species O f the observed A shy-headed and R uddyheaded G eese, 70% and 60%, respectively, occurred in association w ith groups of L esser M agellan G eese. In instances w here A shy-headed G eese (A ) occurred w ith L esser M agellan G eese (L ), the num bers of the tw o species in the flocks w ere correlated positively (n A = n L , r= 0.8 9, d f= 8, PC 0.001). In all th ree regions, m ost groups of the L esser M agellan G oose w ere not associated with dom estic anim als (T able 5). T o clarify this relationship, the incidence of dom estic livestock was n o ted betw een P unta D elgada and P unta A renas, and in T ierra del Fuego. T he follow ing g ro u p freq u en cies w ere recorded: L esser M agellan G oose only = 173; sheep only = 77; cattle only = 2; geese and sheep = 11; and, geese and cattle = 5. T hese data suggest th at there m ay indeed be a positive association betw een geese and cattle, but the sam ple is too small for Table 5. Percentage frequency of association between groups of the Lesser Magellan Goose and domestic livestock in three regions of southern Patagonia, November Frequency of association No. None With sheep With cattle records North Central South statistical testing. F or sheep, how ever, the observed frequencies o f association are very different from those expected (assum ing in d e p e n d e n c e, X 2= , d f= 2, PcO.OOl), dem onstrating a clear negative relationship betw een the presence o f sheep and L esser M agellan G eese, which m ay be due to avoidance o r subtle differences in habitat requirem ents. T h ere w ere m arked latitudinal differences in the habitats used by the L esser M agellan G oose which reflect habitat availability (T able 6, X 2 =77.57, d f= 4, PcO.OOl). Table 6. Percentage frequency of habitat utilisation by the Lesser Magellan Goose in three regions of southern Patagonia, November Habitat North Central South Grassland Scrub W ater associated 30 21) 8 No. records Discussion Lesser M agellan G eese in Chile lay eggs in N ovem ber (Johnson 1965), therefore the m id-n ovem ber surveys should have coincided w ith the incubation phase. H ow ever, although m ore than half o f the birds w ere p aired, single m ales assum ed to have incubating m ates hidden from view com

5 L esser M agellan geese in Chile 19 prised only 5% of all m ales recorded, and only six pairs w ere recorded w ith small young. T hus, while no com m ent can be m ade on the proportion of breeding birds in the populations surveyed, it can be assum ed th at data on group size, sex ratio, abundance and geographical distribution are relatively free of biases which might have occurred if a large proportion of the population was breeding at th e tim e of the surveys. Relative abundance T he inform ation on the relative abundance of the R uddy-headed, A shy-headed and Lesser M agellan G eese in southern P atagonia presents a p attern sim ilar to those rep o rted in the m ore recen t literatu re (H um phrey et al. 1970). In effect, the L esser M agellan G oose outnum bers its two congeners by tw o orders of m agnitude, with the R uddy-headed G oose now occurring with som e regularity only in T ierra del Fuego and southern B uenos A ires P rovince. In T ierra del Fuego, o u r survey included nine R uddy-headed G eese out of a total of 775 sheldgeese, accounting for 1.2 %, in com parison to < 0. 1 % ( 6 in 6000) recorded in 1973 (K ing 1981). A fter an initial increase at th e turn of the century, which m ade it one of the com m onest sheldgeese in northern T ierra del Fuego (Scott 1954), the R uddy-headed G oose decreased dram atically after the 1950s to an estim ated population o f few er than 1,000 birds on T ierra de Fuego in the late 1970s (R um boll 1975; King 1981). The po p ulation has been classified as e n dangered" (K ing 1981), w ith th e m ain factors for its decrease being given as: persecution by m an (egg destruction and shooting); predation by the introduced Patagonian G rey Fox D usicyon griseus', and, com petition with dom estic sheep for forage. T he allegation th at sheep deprive the sheldgeese of forage is unsubstantiated (Sum m ers 1985). C attle w ere introduced into southern Patagonia during the m iddle of the nineteenth century and sheep som e 25 years later. T he extensive cattle industry declined tow ards the end of the century and, sheep ranching expanded to cover m ost of the area occupied today. D uring these changes, forests w ere cleared, burned and converted to grasslands. W e hypothesise that the com bination o f forest clearing (that continues today) and cattle ranching favoured an initial increase o f th e R uddy-headed G oose. T he L esser M agellan G oose p re sum ably benefited through these actions as well. It is suggested, how ever, th at the A shy-headed G oose m ay have decreased with the destruction of th e forest, since it appears th at the species breeds m ainly in forest clearings th at are associated with ponds and o th er w etlands (Johnson 1965; H um phrey et al. 1970, pers. obs.). The short cropping of vegetation by the sheep (C raw shay 1907; Sum m ers and D unnet 1984), and perhaps by rabbits (Jaksic and Y anez 1983), possibly initially benefited tw o of the three species of sheldgoose. H ow ever, it m ay also have resulted in the ground becom ing drier and changes occurring in the com position o f the grassy vegetation as well as an encroachm ent o f semiarid adapted scrub at th e expense of grass cover. It w ould follow th at a change in both the quantity and quality o f the grass cover, and not com petition w ith sheep for forage (Sum m ers and G rieve 1982), has been a m ajor factor contributing to the recent decrease of the R uddy-headed G oose in southern Patagonia. O th e r m ajor factors are undoubtedly persecution and the in tro duction of predators. T he L esser M agellan G oose m ay be less selective in its feeding and nesting requirem ents than the R uddyheaded G oose and, consequently, has not (yet) been so affected by these relatively recent changes in h abitat. Sum m ers and G rieve (1982) found th at G reater M agellan G eese in the Falkland Islands consum ed a greater diversity of vegetable m aterial than did R uddy-headed G eese, but sam ple sizes for the latter w ere sm all. A lien m eadow - grasses Poa spp. dom inated the diets of both species, and abundance of Poa was related to the activities o f sheep. Plum age variation T he occurrence of m ale L esser M agellan G eese in tw o colour phases, w hite versus barred, and the separation of these phases along a north-south axis has been know n for a long tim e (D elacour 1954; H um phrey et al. 1970). It is som ew hat surprising, how ever, th at there has been little investigation o f its evolutionary origin and of the factors responsible for th e phenom enon. G reater M agellan G oose fem ales apparently do not norm ally breed until they are two years old, and m ales usually delay onset

6 20 W.R. Siegfried, P.A.R. H ockey, P.G. Ryan an d A.L. Bosnian of breeding until they are three years old (S u m m ers 1983a). C o m p a ra b le in fo r m ation is not available for the L esser M a g e lla n G o o s e. T h e p re s e n t d a ta, how ever, d em onstrate th at birds of all three plum age types belonging to both sexes occurred as apparently m ated m em bers of pairs, and as paren ts of broods. C learly, th en, we have a situation either of variation in fully adult plum age (true colour polym orphism ) o r of certain individuals being able to breed successfully while not yet having full definitive adult plum age. It is necessary to provide a caveat here in adm itting that our ratios o f fem ale plum age types might not be free of e rro r, since varying conditions of light intensity could have affected our judgem ent in distinguishing betw een shades of head colour, and head colour itself m ay change betw een m oults, which can be irregular (Sum m ers 1983b). B ased on th e data for group size, and incidence of pairs and parents w ith broods (T able 2), it appears th at breeding started earlier in th e north-w estern than in the south-eastern areas of southern Patagonia. T he north-w estern area is generally m ore m esic, w ith relatively m any sm all lakes and o th e r w etlands, and is m ore varied topographically th an th e so u th -eastern area w hich has a longer established and m ore intensive system o f dom estic livestock grazing. D oes this suggest th at th e n o rth w estern area, in which w hite-phase m ales predom inate, provides relatively superior breeding habitat for the L esser M agellan G oose? T his question cannot be answ ered at p resent, but existing ecological theory w ould be satisfied if the L esser M agellan G o o se 's dense n o rth -w estern b reed in g population consisted m ainly of older experienced birds exploiting optim al h abitat, causing younger subordinate individuals to travel farth er south w here habitat is less suitable and breeding occurs later. T his, of course, begs the question o f age-related plum age types, particularly in the case of fem ales. N otw ithstanding th e preceding suggestions and qualifications, it w ould seem that the L esser M agellan G oose does indeed exhibit tw o, or m ore, definitive colour phases, at least in fully adult m ales. The d ata, thus, serve to reaffirm a previously noted tendency for pure w hite m ales to predom inate in the north and barred m ales in the south, of southern Patagonia (H um phrey et al. 1970). T he data also point, for th e first tim e, to a d e c re a s e in th e interm ediate-plum age type in accord with the north to south gradient. T he situation in N ovem ber 1985 was som ew hat different to th at in M arch 1953 (Scott 1954) when w hite m ales accounted for alm ost 50% of males on the m ainland around P unta A renas, but well u nder 1% in T ierra del Fuego. T here is also a segregation of plum age types over a com parably narrow latitudinal range on the w intering grounds in A rgentina (M artin et al. 1986), w ith barred birds dom inating along the coast, and w hite birds dom inating inland to the north. B earing in m ind o u r reservation concerning headcolour, o u r data suggest th at th ere might be a north-south cline in the distribution of at least greyish cinnam on fem ales on the breeding grounds as well. G iven the state o f our know ledge at present, we cannot im prove on the speculation of H um phrey et al. (1970) th at, p rior to the arrival of E uropean settlers and sheep, the tw o colour phases of the L esser M agellan G oose might have been... m ore separated ecologically during the breeding season and that this separation m ay have dim inished during th e past 75 o r 100 years because o f changes in th e e n v iro n m ent b ro u g h t ab o u t by sh eep. Acknowledgements This paper is based on part of the FitzPatrick Institute s 25th Anniversary Expedition to Chile. The work was supported by the South African CSIR and the University of Cape Town. We thank D r Janet Kear and an anonymous referee for their comments on the manuscript. Summary Sex ratios in sheldgeese, especially the Lesser Magellan Goose Chloëphaga p. picta were determined from roadside counts over 800 km in southern Chile and on Tierra del Fuego. They were close to parity. Males with white underparts predom inated in the north and those with barred underparts in the south. There may also be a north/south cline in the greyish-headed females versus those with reddish heads. The interactions with other species of sheldgeese and with cattle and sheep are discussed.

7 Lesser Magellan geese in Chile 2! References Blake, E.R Manual o f Neotropical Birds. Vol. 1. Univ. Chicago Press, Chicago. Crawshay, R The Birds o f Tierra del Fuego. Bernard Q uaritch, London. Delacour, J Waterfowl o f the World. Vol. 1. Country Life Ltd., London. Hum phrey, P.S., Bridge, D., Reynolds P.W. and Peterson. R.T Birds o f Isla Grande (Tierra del Fuego). Smithsonian Institution, W ashington, D.C. Jaksic, F.M. and Yanez, J.L Rabbit and fox introductions in Tierra del Fuego: history and assessment of the attem pts at biological control of the rabbit infestation. Biol. Conserv. 26: Johnson, A.W The Birds o f Chile and Adjacent s o f Argentine. Bolivia, and Peru. Vol. 1. Platt Establecimientos Gráficos S.A., Buenos Aires. King, W.B Endangered Birds o f the World: the IC BP Bird Red Data Book. Smithsonian Institution. W ashington, D.C. Martin, S.I., Tracanna, N. and Summers, R.W Distribution and habitat use by sheldgeese populations wintering in Buenos Aires Province, Argentina. W ildfowl 37: Rumboll, M.A.E E l Cauguen de Cabeza Colorada (Chloëphaga rubidiceps)', una nota de alarma. E l Hornero 11: Scott, P South America W ildfowl Trust. Ann. Rep. 5: Summers, R.W. 1983a. The life cycle of the Upland Goose Chloëphaga picta in the Falkland Islands. Ibis 125: Summers, R.W. 1983b. M oult-skipping by Upland Geese Chloëphaga picta in the Falkland Islands. Ibis 125: Summers, R.W The size and composition of sheld-geese populations and their food consumption on different vegetation types in the Falkland Islands. J. Appi. Ecol. 22:1-17. Summers, R.W. and D unnet, G.M Sheld-geese and man in the Falkland Islands. Biol. Conserv. 30: Summers, R.W. and Grieve, A D iet, feeding behaviour and food intake of the Upland Goose (Chloëphaga picta) and Ruddy-headed Goose (C. rubidiceps) in the Falkland Islands. J. Appi. Ecol. 19: Prof. W.R. Siegfried, Dr P.A.R. Hockey, Peter G. Ryan and Dr Alison L. Bosnian. FitzPatrick Institute, University of Cape Town, Rondebosch 7700, South Africa.