The food habits of Greylag and Barheaded Geese in the Keoladeo National Park, India

Transcription

1 The food habits of Greylag and Barheaded Geese in the Keoladeo National Park, India B E T H A. M ID D L E T O N and A.G. V A N D E R V A L K In troduction T h e fo o d h a b its o f G re y la g and B a rh e a d e d G eese have never been q u an M o re th a n 6,000 geese overw inter in the K eo lad eo N atio n al P ark, B h aratp u r, R a ja s th a n, India. T h e E astern G reylag G oose A n se r anser rubrirostris is the m ost com m on go o se, w ith a p eak population of about tified fo r any p art o f th eir overw intering gro u n d s on th e subco n tin en t of India. Som e factors co n trib u tin g to seasonal food selection (i.e. availability versus physiological needs fo r m igration and breeding) also need 5,000 birds. T hey arrive in the P ark by early to be ex p lo red. N o v em b er and leave by m id M arch, a p a tte rn w hich is typical of n o rth ern India (A li and R ipley 1968). T he B arheaded G o o se A n se r indicus at betw een 500-1,000 individuals is less num erous. T hey arrive in num b ers by late D ecem b er and leave by m id A pril. T his goose has undergone a significant p o p u latio n decline in the past several decades (G o le 1982). It w as n o t possible to kill geese in a N atio n al P ark, so faecal analysis w as used, a m eth o d w hich enables th e population to be resam p led th ro u g h o u t th e season (S torr 1961; S tew art 1967). G eese digest relatively little o f th e plan t m aterial ingested, com p a re d to o th e r v erteb rate herbivores (M attocks 1971 ; O w en 1975). T hey digest about 28% o f th e cellulose and 25% of the hem i- Figure 1. Location of study site, Keoladeo National Park, Bharatpur, Rajasthan, India (upper left). A reas grazed by goose flocks of m ore than 100 birds, Novem ber 1985-April 86 (lower right). Shaded are Greylag goose areas, darkened are Barheaded areas abandoned by Greylags. Letters correspond to area (block) names. 94 W ildfowl 38 (19X7):

2 Food habits o f Indian geese 95 celluloses in th eir diets (B uchsbaum et al. 1986). U n d ig ested food can pass through a goose in as little as one h o u r (O w en 1975; E b b in g e et al. 1975; D orozynska 1962). In sh o rt, it ap p e a re d th a t the food habits of g e e se c o u ld b e d e s c rib e d a d e q u a te ly th ro u g h faecal analysis. M ethods S tu d y site T h e K eo lad eo N ational P ark (27 13'N I T 3 2 'E ) n ear B h a ra tp u r, 50 km w est o f A gra (F igure 1), lies in th e floodplain of the G am b h ir and B anganga R ivers. This w etland lies in a n atu ral d epression, but each y ear th e w ater d ep th is en h an ced by adding rain w ater th ro u g h a floodgate from a reserv oir, the A ju n B und. T he p ark is u n d er the co n tro l o f th e S tate G o v ern m en t of R ajasth an F o restry D ep artm en t. P rio r to 1973, w hen it w as d eclared a w ildlife sanctuary, th e area w as used as the M aharajah of B h a ra tp u r s p rivate shooting reserve (A li an d V ijayan 1986). W a te r levels in this w etland fluctuate b o th w ithin and betw een years. A fter the m onsoon (Ju n e -S e p te m b c r), w ater levels are at th e ir annual m axim um. A s th e m onsoon w aters recede during the autum n and w in ter, p arts o r all m ay becom e dry (G opal 1973). A fte r a p o o r m onsoon, the w hole w etlan d m ay be dry fo r th e en tire year. In a very heavy m onsoon year, the w ater m ay be m o re th a n tw o m etres deep, w ith m ost of th e P a rk s 29 km 2 area in u n d ated for at least o n e m onth. T his w etland is d o m in ated by one species o f grass, P aspalum distichum (M aheshw ari 1963). T his becam e m uch m ore prevalent a fte r th e banning o f cattle and buffalo grazing in 1981 (A li and V ijayan 1986). O nly th e d eep est p arts o f th e w etland arc now free o f this grass; these are dom inated by su b m erg ed plan t species such as P otam o geton in d icu s and H ydrilla verticillata. S avanna areas w hich ten d to be flooded for only sh o rt p eriods after a heavy m onsoon are d o m in ated by th e tall grass species, Vetiveria ziza n io id es, w ith occasional trees, prim arily A cacia nilotica. G eese, m ost typically fo u n d in th e P. distichum dom inated a reas, have also been recorded in the o th er tw o v egetation types in B h aratp u r. In V. ziza n io id es stands, they ten d to be found in th e w e tte r depressions th a t are free of tall grasses. Collection o f G oose D roppings and their analysis G oose droppings w ere collected every m onth from w herever goose flocks w ere currently grazing. T o ensure a good sam pling over the population. 100 fresh d ro p pings w ere collected over a w ide area in a given location. G reylag G oose droppings w e re d is tin g u is h e d fro m th o s e o f B arheaded G eese on size, colour, texture and location. Typically, the tw o goose species prefer to graze in separate but adjacent flocks in a single area (Figure 1). T he 100 droppings w ere taken to the laboratory, carefully cleaned of debris, and then m ixed. This com posite sam ple was hom ogenized in an electric blender on the sam e day th at it had been collected in the field. H o m o g e n iz a tio n im p ro v e d th e quality of the slides by clearing dark pigm ents from the m aterial. A fter m ixing for one m inute, the m aterial was w ashed for about five m inutes through a 0.1 mm (200 m esh) screen until thoroughly cleaned. This cleaned m aterial was dried in an oven at 72 C for future slide prep aratio n or im m ediately m ounted on slides (Scott and D ahl 1980). Five duplicate slides w ere prep ared for each m onthly com posite sam ple (n) follow ing a m ethod described by B aum gartner and M artin (1939). A sm all am ount of faecal m aterial was boiled in a drop of H ertw ig s solution over a sm all burner. N ext, a few drops of H o y er s m ounting m edium w ere put on th e boiled m aterial, heated and covered with a 22 x 50 mm cover slip while still hot. T he m aterial was spread evenly over th e slide to m inim ise fragm ent overlap. C om pleted slides w ere then dried at room tem p eratu re o r a t 55 C in a drying oven. This m ethod does not w ork during hum id w e a th e r c o n d itio n s due to the h y d ro p h ilo u s n a tu re o f th e m o u n tin g m edium. A reference plant collection for the p u r pose of identifying faecal fragm ents was c o n stru c te d. S am ples o f know n p lant species w ere first chopped, then ground in a blender follow ed by the sam e cleaning and m ounting m ethods used for the faecal m aterial. B ecause leaves, stem s, seeds and roots arc readily distinguishable from each

3 96 Beth A. M iddleton and A. G. Van Der Valk o th er, a reference slide was m ade for each plant part. C ertain species changed slightly in m icroscopic appearance during the season (e.g. trichom es in P. distichum ) so several series o f slides w ere prep ared as necessary. Slide reading follow ed th e sequence described by Scott and D ahl (1980). S tarting at th e upper left hand corner o f the slide, each fragm ent was recorded in a m icroscope field w here three identifiable fragm ents occurred until 50 fields had been read. From this inform ation, percentages for each slide w ere obtained. F or both goose species, five duplicate slides w ere read and averaged for each m onthly com posite, in each goose grazing location with m ore than 100 geese. T he com ponents of variance for the m eans of the plant fragm ents in the goose diet w ere analysed using the Proc V arcom p option in SAS (Statistical A nalysis System 1982). T he m ethod provides estim ates in term s of surface area only. Since the relationship betw een area and volum e or w eight varies w idely am ong plant parts, the results give only an approxim ate m easure o f the contribution o f the various plants to the diet. Results G reylag Diet T he main source of variation in food habits derives from th e m onth collected rather than goose feeding location o r duplicate slides (T able 1). Seasonal diet differences are apparent. E arly in the overw intering season, seeds are com m on in the diet of G reylag G eese. Seed ripening occurs for m any plants during the postm onsoon con- Table 1. Components of variance of selected plant species in the diets of Greylag and Barheaded Geese. Variance Component Estimate Month Area Sample G R E Y L A G G O O SE Paspalum distichum Nymphaea nouchali Oryza rufipogon B A R H E A D E D G O O SE Paspalum distichum Pisum sativum ditions in N ovem ber and D ecem ber and seeds m ade up 64.7% and 58.4% of the diet in these m onths. M ost of these seeds are from N ym phaea nouchali, O ryza rufipogon, N ym phoides cristatum, and P. distichum. W hen seeds becom e scarce (they contributed only 1.5% in January and less subsequently), grasses increase in im portance, to form alm ost all of the diet (Figure 2) from January until m igration. M ost of the grass contribution com es from the stem s and leaves o f the one species, P. distichum, 68% over th e entire season. In all, at least 21 plant species are eaten by G reylag G eese, w ith additional, m inor contributions from insects and snails (T able 2). B arheaded G oose Diet Striking differences are seen in the seasonal eating pattern s of this species (Figure 2 and T able 3). G rasses are im portant in January and F ebruary but drop sharply in M arch and A pril w hen herbaceous plants and seeds becom e im portant. In January, the main constituent of the diet is W heat Triticum aestivum (T able 3). In F ebruary, when field crops are not available, native P. distichum predom inates. O verall, its leaves com prise 45% of the overw intering diet. By M arch and A pril, once again th e diet is m ostly field crops, but chiefly of Peas Pisum sativum and G ram s Cicer arietinum instead of w heat. Seeds then m ake up 41.9% and 73.3% of th e diet, having been insignificant in th e tw o earlier m onths. B arh ead ed G eese utilised a total of 21 different plant species, 3 o f which w ere o f agricultural origin. Insects and snails constitute a very low percentage of th e diet throughout the season (T able 3). B arheaded G eese w ere thus less dep en dent than G reylag G eese on native w etland plants and only in February did these com prise all their diet, 95% of it being P. d is tic h u m. In p a rts o f th e s e a s o n. B arheaded G eese relied on agricultural crops. O verall, they utilised 52% wild plants in th eir diet, w hereas G reylag G eese used nothing but wild plants. D iscussion G eese feed on a wide variety of plants (M cfarland and G eorge 1966; Pollard and

4 Food habits o f Indian geese 97 n=1 n=2 n-2 MONTH OF THE YEAR n=1 n=2 n=4 n =5 n=5, MONTH OF THE YEAR Figure 2. Monthly comparison of fragment type in the faeces of Greylag (bottom) and Barheaded Geese (top), November 1985-March 86, in the Keoladeo National Park, Bharatpur, Rajasthan, India. Num ber of composite samples from goose flocks (n).

5 98 Beth A. M iddleton and A. G. Van Der Valk Table 2. Pood habits of the Greylag Goose, Keoladeo National Park, Bharatpur, India, Averages for all plant species in diet, given by percent. This percent divided between leaves, stems, seeds (in parenthesis). * = < 0. 1%. Plant Nov Dec MONTH Jan Feb Mar Average Paspalum distichum leaves (8.4) (1.9) (33.4) (11.7) (13.5) (13.8) stems (8.4) (17.3) (55.5) (80.4) (82.2) (48.7) seeds ( 12.6 ) (11.9) ( 1. 1) (0.2 ) (0.4) (5.2) Nymphaea nouchali leaves (3.0) (0.8 ) (0.8) seeds (52.1) (8.6 ) * - - ( 12. 1) Oryza rufipogon leaves ( 1.2) ( 11.6 ) (2. 1) (0.9) (0.8 ) (3.3) seeds - (35.1) (0.2 ) - (0. 1) (7.1) Ipomoea aquatica leaves ( 1.2) (0.8 ) (0.9) (2.8 ) (0.5) ( 1.2) seeds - (0. 1) (0. 1) Sagittaria guayanensis «1.0 leaves (2.4) ( 1.8 ) n - - (0.8 ) seeds - (0.7) (0. 1) N ymphoides cristatum * leaves (7.8) (0.3) ( 1.6 ) seeds - (1.5) (0. 1) (*) (0. 1) (0.3) Scirpus tuberosus Cynodon dactylon Panicum paludosum (stem) _ grass seed Paspalidium flavidum Sporobolus sp herbaceous seed * H em iadelphus polyspermus 0.1 * 0.1 * «Scirpus littoralis * 0.4 Eleocharis palustris _ Utricularia sp * - Ceratophyllum demersum * 0.1 _ * Cyperus alopecuroides 0.3 * _ 0.1 grass stem * * Insects & snails W altcrs-d avies 1968; O w en and K erbes 1971; O w en 1976; B urton et al. 1979; Prevett et al. 1979; Sum m ers and G rieve 1982). T he particular spccics of plants consum ed depends on the flock s geographic location and on plant availablility. In th e K e o la d e o N a tio n a l P a rk, B arheadcd G eese cat m ore gram ineous leaves than they arc rep o rted to do elsew here. In th eir nesting range in th e Soviet

6 Food habits o f Indian geese 99 Table 3. Foods habits of the Barheaded Goose, Keoladeo National Park, Bharatpur, India, Averages for all plant specics in diet, given by percent. This percent divided between leaves, stems, seeds (in parenthesis). * < 0. 1%. Jan MONTH Feb Mar Apr Average Paspalum distichum leaves (10.3) (94.y) (49.5) (26.4) (45.3) stems seeds - (0. 1) - - n Pisum sativum seed - - (42.2) (72.1) (28.5) stems & leaves (4.6) - (2.3) (0.2 ) ( 1.8 ) Cicer arietinum seed - - (0. 1) (0.6 ) (0.2 ) stems & leaves (3.1) - (2.6 ) - (1.4) Triticum aestivum Cynodon dactylon Ipomoea aquatica (leaves) 0.<S _ 0.3 Paspalidium flavidum Panicum paludosum _ 0.8 Ceratophyllum demersum Sporobolus sp Scirpus tuberosus 2.6 ( N ymphoides cristatum Scirpus sp * Sagittaria guayanensis (leaves) Scirpus littoralis 0.1 * Eleocharis palustris 0.1 * Utricularia sp H emiadelphus polyspermus 0.2 _ * Nymphaea nouchali (seed) 0.2 * herbaceous leaf Insects & Snails U nion, they feed m ostly on terrestrial vegetation. including the seeds o f grasses and legum es; in tidal areas they eat seaw eeds, crustaceans and invertebrates (D em ent ev and G ladkov 1952). Greylag Gecsc in the Kcoladco National Park eat seeds early in the w inter season and grass parts in later m onths, not greatly d ifferen t from rep o rts o f th e E astern G reylag diet in the Soviet U nion. D em ent ev and G ladkov (1968) reported that they eat grasses of various sorts (w inter rye. hair grasses, aw nless broom grass Leerzia sp.), the fruits of Potam ogeton pectinatus and the seeds of Polygonum sp. and oats. In the N etherlands (L oosjes 1974) and in Spain (A m at 1986) G reylag G eese eat a wild plant diet of mostly Scirpus lacustris root pieces

7 100 Beth A. M iddleton and A. G. Van Der Valk and the tubers o r rhizom es o f Scirpus m aritim us. W estern G reylag G eese A. a. anser in Scotland eat mainly grasses but in late sum m er switch to ripening oats (N ew ton and K erbes 1974). In the w inter, they prim arily feed in agricultural fields w here they eat oat and barley seeds, later sw itching to p o tato crops (N ew ton and C am pbell 1973; N ew ton et al. 1974). G reylag G eese in the K eoladco N ational Park have very different food habits from the B arheaded G eese grazing in the.sam e area. L arge-billed species such as the form er arc ad ap te d for ro o tin g, w hereas sm aller-billed geese like the latter m ay be b e tte r suited for grazing on sh o rter grasses (R eed 1976) in shallow er w ater. O th er studies have d e m o n stra te d th a t goose species in one area m ay have very different food habits due to differences in food p re fe re n c e (C ra v e n 1984) as w ell as m orphological differences. T he G reylag G eese m aintain them selves on the nearly apple-sized fruits of N ym phaea nouchali (and m any kinds of seeds) in the early part of th eir overw intering season. L ater in the season, they can be observed forcibly pulling up w hole culm s of P. distichum which may be grow ing in w ater approaching 2 m etres deep. B arheaded G eese which have a m uch sm aller bill, tend to tear off the leaves of grasses rath er than tackle the bulky stem s. T hey can often be seen grazing in shallow areas abandoned by G reylags, perhaps preferring the shorter forage provided. B arheaded G eese spend m ost o f their tim e grazing in agricultural fields during certain parts o f th e w inter season, but during the day, they also graze on wild grasses in the w etland. Since m ore than half of their diet com es from a wild food source, they cannot be thought as being com pletely dep en d en t on agriculture. It is not know n w hether o r not P. distichum was such an im portant food item in the diet of geese p rior to this plan t s spread after the cessation o f cattle and buffalo grazing in It is clear from this study th at it is now an im portant com ponent of the diet for both goose species overw intering in the K eoladeo N ational Park. Studies now in progress will explore the role of geese in creating openings in this nearly m onospecific grass cover. T o som e extent, seasonal dietary requirem ents of geese dictate their choice of food (Y denberg and Prins 1981; H alse 1984; H obaugh 1985). T hese authors suggest that geese choose m ore nutritious food prior to m igration or to breeding in o rd er to accum ulate fat reserves. In India, B arhcaded G eese do behave in this fashion, in that, p rio r to m igrating to th eir H im alayan breeding grounds, they eat such nutrientrich foods as peas, gram s and, a little earlier in the season, young w heat but then, these plants are only available at this tim e. D uring February, betw een the season of young w heat and th at of ripening legum es, the geese retu rn to graze exclusively in the w etland. O n e c a n n o t m ake a case fo r food selection by G reylag G eese on their overw intering grounds related to m igration or breeding requirem ents. Shortly after they arrive for th eir five m onth sojourn, they choose to eat th e nutrient-rich seeds availab le d u rin g th e p o st-m o n so o n p erio d (N ym phaea nouchali, O ryza rufipogon). Just before m igrating from India, they switch to grasses w hen seeds are no longer available. F or G reylag G eese th en, and to som e extent for B arhcaded G eese, availability seem s to be a m ajor determ inant of their seasonal food selection. Acknowledgements This study was carried out under Smithsonian G rant No Wc would like to thank the S tate G o v ern m en t of R ajasth an F o restry Departm ent for their cooperation with this work, including O.P. M athur, V.D. Sharma and Rajan M athur. For help with various technical aspects of this report, we would like to thank Gretchcn Scott of Texas Tech, Ted Bailey of the Iowa State and Craig B. Davis of Ohio State University. Comments on drafts of this paper came from Ncls Lersten, Bill Clark, Erv Klaas and Jerry Sell. Special thanks to the field assistants in this work: O.P. Mudgal, Laxmi Kant Mudgal, Sunil Mudgal, Kriste Ericcson, Kurt Streeb, M arjorie Boustield. Nihal Singh, Gopal Singh, Prakash Singh, Bihari Singh and Soren Singh. Sum m ary The food habits of overwintering Greylag Anser unser rubrirostris and Barheadcd Anser indicus G eese w ere investigated in the K eoladeo National Park. B haratpur, India. Greylag Geese eat mostly seeds of Nymphaea nouchali and O ryza rufipogon in Novem ber and Decem ber but

8 Food habits o f Indian geese 101 shift to native grasses such as Paspalum distichum in January. They eat mostly P. distichum until they leave in M arch. Barheaded Geese eat mostly young wheat in January, native grasses, such as P. distichum in February and change to field peas and grams in March and April. Native w etland plants m ake up most of the diet of Greylag and B arheaded G eese, 100% and 52% respectively. Overall, P. distichum is the m ost im portant species in the diet of these geese, comprising 67.8% of the Greylag and 45.4% of the B arheaded Goose diet. References Ali, S. and Vijayan. V.S Keoladeo National Park ecology study. Summary report I9S0-85. Bombay Natural Flistory Society, Bombay. India. Ali, S. and Ripley, S.D H andbook o f the birds o f India and Pakistan. Volume 1. Oxford University Press, Bombay, India. Am at, J.A Numerical trends, habitat use and activity of Greylag Geese wintering in southwestern Spain. Wildfowl 37: Baumgartner, L.L. and Martin A.C Plant histology as an aid in squirrel food-habit studies. J. Wildl. Manage. 3: Buchsbaum, R., Wilson, J. and Valiela, I Digestibility of plant constituents by Canada Geese and Atlantic Brant. Ecology 67: Burton, B.A., Hudson. R.J., and Bragg, D.D Efficiency of utilization of bulrush rhizomes by Lesser Snow Geese. J. Wildl. Manage. 43: Craven, S.R Food habits of Canada Geese on the coast of Hudson Bay. J. Wildl. Manage. 48: D em ent ev, G.P. and Gladkov, N.A Birds o f the Soviet Union. Vol. 4, Moscow. Dorozynska, N Food intake and defecation in the goose, Anser anser L. Acta Biologiae Experimentalis 3: Ebbinge, B., Canters, K. and D rent, R Foraging routines and estim ated daily food intake in Barnacle Geese wintering in the northern Netherlands. W ildfowl 26:5-19. Gole, P Status of Anser indicus in Asia with special reference to India. Aquila 89: Gopal, B A survey of the Indian studies on ecology and production of wetland and shallow wpter communities. Polskie Archiwum Hydrobiologii 20: Halse, S.A D iet, body condition and gut size of Egyptian Geese. J. Wildl, Manage. 48: Hobaugh, W.C Body condition and nutrition of Snow Geese wintering in southeastern Texas. J. Wildl. Manage. 49: Loosjes, M O ver terreingebruik, verstoringen en voedsel van Grauwe Ganzen Anser anser in een brak getijdengebied. Limosa 47: M aheshwari, J.K The flora o f Dehli. CS1R. New Dehli, India. M attocks, J. G Goose feeding and cellulose digestion Wildfowl 22: M cfarland, L. Z. and George. H Preference of selected grains bv geese. J. Wildl. Manage. 30:9-13. Newton, I. and Campbell, C.R.G Feeding of geese on farmland in east- central Scotland../. Appi. Ecol. 10: Newton, I., Campbell, C.R.G. and Allison. A Gut-analysis of Greylag and Pink-footed Geese. Bird Study 21: Newton, I. and Kerbes, R.H Breeding of Greylag Geese (A nseranser) on the O uter Hebrides. Scotland. J. Anim. Ecol. 43: Owen, M An assessment of fecal analysis technique in waterfowl feeding studies../. Wildl. Manage. 39: Owen, M The selection of winter food by W hite-fronted Geese. J. Appi. Ecol. 13: Owen, M. and Kerbes, R.H On the autumn food of Barnacle Geese at Caerlaverock National Nature Reserve. W ildfow l-22: Pollard, D.F.W. and W alters-davies, P A preliminary study of the feeding of the Greenland W hite-fronted Goose Anser albifrons flavirostris in Cardiganshire. W ildfowl 19: Prevett, J. P., Marshall, I.F. and Thomas. V.G Fall food of Lesser Snow Goose in the Jam es Bay Region. J. Wildl. Manage. 43: Reed, A Geese, nutrition and farmland. Wildfowl 27: Scott, G., and Dahl, B.E Key to selected plant species o f Texas using plant fragments. Occasional Papers, The Museum, Texas Tech University. Lubbock. Texas. Statistical Analysis System SA S user s guide: statistics. Cary, North Carolina.

9 102 Beth A. M iddleton and A. G. Van D er Valk Stewart, D.R.M Analysis of plant epidermis in faeces: a technique for studying the food preferences of grazing herbivores. J. Appi. Ecol. 4: Storr, G.M Microscopic analysis of faeces, a technique for ascertaining the diet of herbivorous mammals. Aust. J. Biol. Sci. 14: Summers, R.W. and Grieve, A D iet, feeding behavior and food intake in the Upland Goose (Chloëphaga picta) and Ruddv-headed Goose (C. rubidiceps) of the Falkland Islands. J. Appi. Ecol. 19: Ydenberg, R.C. and Prins, H.H.Th Spring grazing and the m anipulation of food quality by Barnacle Geese. J. Appi. Eco!. 18: Beth A. Middleton and A.G. van der Valk, Botany Departm ent. Iowa State University Ames, Iowa