402 SHORT COMMUNICATIONS

NELSON,K. S. 1997. Marbled Murrelet (Bruchyram- and enhancementsthrough release 6.11. SAS In- phus marmoratus).In A. Poole and E Gill [eds.], stitute. Inc.. Carv. NC. The of North America, No. 276. The Acad- SEALY,S. G. 1972. Adaptive differences in breeding emy of Natural Sciences, Philadelphia, and The biology in the marine family Alcidae. Ph.D. American Ornithologists’ Union, Washington, diss., Univ. Michigan, Ann Arbor, MI. DC. SEALY.S. G. 1975a. Asuects of the breeding biologv of the Marbled Muielet in British Columbia. Bi;h PITOCCHELLI,J., J. PIATT, AND M. A. CRONIN.1995. Banding 46:141-154. Morphological and genetic divergence among SEALY,S. G. 1975b. Feeding ecology of the Ancient Alaskan populationsof Brachyrumphusmurrelets. and Marbled Murrelets near Langara Island, Brit- Wilson Bull. 107:235-250. ish Columbia. Can. J. Zool. 53:418-433. RODWAY,M. S. 1990. Status report on the Marbled WHITWORTH,D. L., J. Y. TAKEKAWA,H. CARTER,AND Murrelet in Canada. Committee on the Status of W. R. MCIVER. 1997. A night-lighting technique EndangeredWildlife in Canada, Ottawa, Canada. for at-sea capture of Xantus’ Murrelets. Colonial SAS INSTITUTE.1996. SAS/STAT software: changes Waterbirds20:525-53 1.

Thhr Condor 101:402-407 0 The Cooper Ornithological Soc~ty 1999

FACILITATION OF SAP-FEEDING BIRDS BY THE WHITE-FRONTED IN THE MONTE DESERT, ARGENTINA’

PEDROG. BLENDINGER BiodiversityResearch Group, IADIZA-CRICYT, CC 507, 5500 Mendoza, Argentina, e-mail: [email protected]

Abstract. In the Monte desert of South America, fluye de las perforaciones.Otras especiesde aves, de the overall supply of water and food for birds decreas- diferentes grupos ecol6gicos y taxon6micos, aprove- es in the dry, cold season (June through September). than este recurso, que de otro modo raramente esta During this period the White-fronted Woodpecker disponible. La savia es un componenteimportante de (Melanerpes cactorum) drills holes in the trunks and la dieta del Carpintero de 10sCardones y de otras 11 branchesof Prosopisjexuosa and feeds on the exuded especies de aves; durante 10s meses de junio y julio sap. Other bird species,diverse in ecological attributes constituye de1 16% al 83% de las observacionesde and taxonomic affinities, take advantage of this re- forrajeo. La agresi6n por 10s Carpinteros de 10s Car- sourcewhich otherwise would be rarely available. Sap dones reduce significativamenteel tiempo de alimen- is a major constituentof the diet of the White-fronted tacion en las perforaciones con savia de las especies Woodpeckerand 1I other bird species,and sap feeding m&s pequeiias. En la explotacidn de este recurso se comprises between 16% to 83% of foraging observa- revela una compleja combination de interacciones tions made during June and July. Aggression by positivas y negativasentre las aves. Las interacciones White-fronted significantly reduced the agonisticassugieren que el Carpintero de 10sCardones time smaller bird species spent feeding on sap, indi- compite activamentepor este recurso,mientras que las cating that White-fronted Woodpeckersactively com- otras especies de aves se favorecen al acceder a un pete for this resource. Other bird species profit from recurs0 rice en agua y en azucares. having accessto a resourcerich in water and sugar. Key words: Argentina, competition,facilitation, In Europe and North America, several species of Melanerpes cactorum, Monte desert, sap feeding, woodpeckers feed on the sap flowing from drilled White-frontedWoodpecker. holes (Foster and Tate 1966, MacRoberts and Mac- Roberts 1976, Short 1982). This behavior is perhaps Resumen. En el desiertode1 Monte de Sudamerica, best exemplified by the Yellow-bellied Sapsucker la oferta de agua y aliment0 para las aves disminuye (Sphyrupicusvarius). The sap from holes drilled by durante la estacion seca de inviemo Cjunioa septiem- this speciesattracts, and is used by, other taxa includ- bre). Durante este periodo, el Carpintero de 10s Car- ing insects,birds, and mammals (Foster and Tate 1966, dones (Melanerpes cactorum)taladra 10stroncos y ra- Wiens 1989, Holmes 1990). Use of this food resource mas de Prosopisjexuosa y se alimenta de la savia que may be particularly important to hummingbirds, be- cause they require food of high energy content (Miller and Nero 1983). ’ ’ Received 26 January 1998. Accepted 30 Septem- In South America, two species of woodpeckersof ber 1998. the Melanerpes use sap as a food resource:the SHORT COMMUNICATIONS 403

Acorn Woodpecker (M. formicivorus) of the lower of 13 nets were operated periodically, each of them montane wet forest in Colombia (Kattan and Murcia distributed on an area of about 7 ha. 1985, Kattan 1988) and the White-fronted Woodpecker (M. cactorum) (Genise et al. 1993). In the thorn-scrub RESULTS woodland of the semiarid Chaco in Argentina, the During the dry season, White-fronted Woodpeckers White-fronted Woodpecker feeds on sap only during drill holes through the bark and into the phloem of the austral winter; other avian species and insects are Prosopis jenuosu trunks (Fig. 1). Either individually attracted to the drilled holes (Genise et al. 1993). Here or in flocks of up to six birds, White-fronted Wood- I document sap consumption by White-fronted Wood- peckers periodically visited the drillings to feed on the peckers in woodlands occupying the driest portion of exuded sap. Each group of woodpeckers simultaneous- their range, interactions with other bird species that use ly maintained numerous trees with active holes. On a this resource, and seasonal variation in the exploitation few occasions, I observed holes with flowing sap in of sap. If sap is a valuable resource during the dry trunks of Bulnesiu retumu, whereas the two other tree season when both water and food are scarce, then inter- species present in the area, Geqfroea decorticuns and intraspecific agonistic interactions are expected. (chahar) and Ephedru boelckei (patron), were not used by woodpeckers. During the focus sampling period METHODS (June-July), woodpeckers drilled holes only in trunks The study was conducted from November 1996 and branches of P. j?exuosu. I observed only one sap through November 1997 in the Flora and Fauna Re- hole in a trunk of B. retumu in the month of June. serve of Telteca (32”21’S, 68”03’W), Mendoza Prov- However, subsequent observations suggest that, as the ince, Argentina. This reserve is located within the dry season progressed, woodpeckers modified their Monte Phytogeographical Province (Cabrera and Wil- foraging strategy. By early September, they continued link 1980), a narrow arid and semi-arid latitudinal strip to use P. ,jIexuosu, but B. retumu and Lurreu divaricutu in western Argentina, at the foot of the Andes. Telteca shrubs also were frequently drilled, and occasionally is a desert area with marked climatic seasonality, rainy Cupparis atumisqueu. By September, only 33% of the summers, and cold dry winters. Mean annual rainfall observations of sap feeding by all bird species was of is 161 mm (data from Meteorological Program of the P. flexuosa, and 30% each of B. retuma and L. divur- Regional Center for Scientific and Technological In- icatu. vestigations, Mendoza). The landscape is characterized The number of holes in trees and the frequency of by a system of dunes and scrublands dominated by feeding events of the White-fronted Woodpecker Larreu divaricatu Cjarilla), Trichomuriu usillo (usillo), were positively correlated (r3 = 0.65, n = 11, P = and Bulnesiu retamn (retamo). Open woodlands of 0.03). Other bird species clearly took advantage of Prosopis$exuosu (algarrobo dulce) grow in the low- sap flow (Table l), only accessible at recently drilled lying areas between dunes, with a shrub layer domi- woodpecker holes. It is not surprising that the fre- nated by B. retumu, Cuppuris utumisqueu (atamisqui), quency of visits by the other birds also was higher at and Lycium tenuispinosum (Ilaullfn). the most intensely drilled trees (v, = 0.78, n = 11, P In June and July 1997, I made three to nine (mean = 0.005), regardless of the presence or absence of 2 SD = 7.4 ? 2.5) 5-min focal observations of all woodpeckers. birds visiting 11 P. flexuosu trees that were being used Sap consumption by birds was inversely correlated by eight White-fronted Woodpecker flocks. I also re- with rainfall (Fig. 1). During June and July, for bird corded the number of individuals of each species that species using this resource, at least 16% of my for- fed on sap, and the time they spent at the sap tree. aging records was comprised of sap from drilled From November 1996 through October 1997, I made holes (Fig. 2). For each species, I compared the fre- periodic random observations of the foraging behavior quency of feeding at focal sap trees with the species’ of birds showing sap-feeding habits. Whenever possi- relative abundance estimated with mist-nets (Table ble, I collected information on feeding behavior. To 1). Sap consumption was not proportional to bird spe- ensure sample independence, I recorded a single feed- cies abundance in the Telteca Reserve (Fig. 3). At ing event for each individual and, when flocks were least five of these species appeared to actively select involved, only one individual per species (Hejl et al. this resource, including the White-fronted Woodpeck- 1990, Recher and Gebski 1990). I did not knowingly er and Tufted Tit-Spinetail (Leptusthenuru platensis), include the same individual twice in any one day. In whereas other abundant species such as the Common order to diminish overestimation of the most conspic- Diuca-Finch (Diucu diucu) and Picui Ground-Dove uous foraging activities, I only considered the feeding (Columbinu phi) were not observed to use it at all behavior 10 set after a focal bird had been sighted (Fig. 3). (Wiens 1989). White-fronted Woodpeckers defend their feeding The number of sap holes in each tree was estimated holes by driving away smaller bird species attempting by counting the number of trunks and branches, over to use them. On 58.2% of the 122 occasions in which 10 mm in diameter, showing drilled holes with actual I observed smaller birds (weight < 11 g) like the sap-flow. Tufted Tit-Spinetail, the Ringed Warbling-Finch I used 35-mm mesh mist-nets, 12.4 m long, to assess (Poospiza torquuta), and the Greater Wagtail-Tyrant bird relative abundance. Nets were placed near shrubs (Stigmaturu budytoides) visiting sap trees with wood- and trees in the low-lying areas between dunes. Mist- peckers present, the smaller heterospecifics were nets were operated between April and August of 1997, chased away by the woodpeckers. However, on only for a total trapping effort of 1,142 net hours. Two sets 6.3% of the observed instances (n = 32) did wood- 404 SHORT COMMUNICATIONS

White-fronted Woodpecker Other species 90

50 60

70 a-F

40 60 rY

% T 50 Y, g 30 E = 40 .o c ii .r 30 2 d 2o % 20 .g

IO 10 s‘ 0* 0 0 JFMAMJJASOND FIGURE 1. Mean monthly rainfall at Encon (30 km northeast of Telteca), and relative occurrence of sap feeding (37 1 foraging-independentobservations from May to September,and 150 from October throughMarch) by White-fronted Woodpecker and other birds (11 species, see Table 1). Correlation between sap consumption and mean monthly rainfall: White-fronted Woodpecker: r, = -0.81, n = 8, P = 0.014; other species: r, = -0.87, n = 8, P = 0.005.

0 Sap feeding n Other foraging behaviors m Monk Parakeet (6)

m White-fronted Woodpecker (21)

_ Tufted Tit-Spinetail (35)

m Ringed Warbling-Finch (23)

_ Golden-billed Saltator (IO)

_ Brow Cacholote (6) _ Greater Wagtail-Tyrant (62)

_ Stripecrowned Spinetail (7)

_ Rufous-collared Sparrow (16) Narrow-billed Woodcreeper (11)

Checkered Woodpecker (6)

-1 Many-colored Chaco-Finch (6) / 1 0.0 0.2 0.4 0.6 0.8 1.0 Proportionof foraging observations

FIGURE 2. Relative importance of sap feeding at Prosopis jexuosa trees in June and July by birds of the Monte desert, Argentina. Number of foraging-independentobservations in parentheses(n = 209). SHORT COMMUNICATIONS 405

TABLE I. Birds feeding on sap recorded at the Telteca Reserve. Columns indicate: relative abundance of each species based on total mist-net captures (April through August, n = 242), total number of individuals per species (n = 323) recorded sap-feeding at White-fronted Woodpecker’s holes in 5-min focal observations, and mean foraging time spent at trees with sap holes.

Mean t SD mnr rpent Body mass (g) Mist-net captures Sap feeding at sap trees (min) White-fronted Woodpecker (Melanerpes cactorum) 34.6 10 118 6.1 2 6.5 Tufted Tit-Spinetail (Leptasthenuru platensis) 9.4 18 61 2.3 t 2.3 Greater Wagtail-Tyrant (Stigmutura budytoides) 10.1 31 60 1.8 i 1.3 Monk Parakeet (Myiopsittu monachus) 94.0 3 32 11.7 ‘-c 9.3 Ringed Warbling-Finch (Poospiza torquata) 10.3 14 30 2.1 k 2.0 Golden-billed Saltator (Saltutoruuruntiirostris) 47.0 6 9 3.3 2 2.9 Many-colored Chaco-Finch (Saltatricula multicolor) 22.9 16 4 4.0 k 2.6 Brown Cacholote (Pseudoueisura lophozes) 64.0 1 3 8.0 i 5.0 Rufous-collared Sparrow (Zonotrichia cape&s) 18.2 38 3 Narrow-billed Woodcreeper (Lepidocolaptes angustirostris) 25.3 6 1 - Checkered Woodpecker ( m&us) 28.2 3 1 Stripe-crowned Spinetail (Cmnioleuca pyrrhophiu) 11.0 4 1 Common Diuca-Finch (Diuca diuca) 25.6 35 0 Picui Ground-Dove (Columbina picui) 45.0 17 0 Short-billed Canastero (Asthenes baeri) 17.4 10 0 Chaco Earthcreeper (Upucerthia certhioides) 22.6 8 0 Gray-bellied Shrike-Tyrant (Agriornis microptera) 52.7 3 0 Least Shrike-Tyrant (Agriornis murina) 27.5 3 0 Crested Gallito (Rhinocrypta lunceolata) 55.0 3 0 House Wren (Troglodytes aedon) 9.5 3 0 Other species (n = 7) 9 0

peckers displace species weighing more than 40 g. DISCUSSION Indeed, there was a clear inverse relation between In the northern Monte desert, the supply of food re- mean body mass of each bird species and the fre- sources strongly decreases during the dry season (Fig. quency of aggressive encounters between them and individual M. cuctorum (r, = -0.96, n = 7, P < 1; Blendinger, unpubl. data). Between May and Sep- 0.001). The observation that smallest species spend tember, the Telteca Reserve receives only approxi- less time on foraging sites (Table I) (r> = 0.86, n = mately 10% of its total annual rainfall. The productiv- 8, P = 0.007) is thus at least partly explained by the ity of drylands is related to the quantity of precipita- agonistic behavior of woodpeckers. Supporting this tion (Dunning and Brown 1982, Wiens 1991). Accord- is the fact that, for all three of the smallest bird spe- ingly, the drastically diminished availability of cies combined, the average time spent at the drilled arthropods and fleshy fruits results in increased short- trees was significantly shorter when woodpeckers age of preformed water for birds and many other or- were present (Mann-Whitney U-test, z = 5.05, n, = ganisms. As the dry season progresses, birds increas- 106, n, = 90, P < 0.001). Although with lower fre- ingly resort to resources (rarely if ever consumed in quency, I also recorded aggressive behavior among summer) which are high in water content. Almost all other bird species, even against woodpeckers. of the bird species < 100 g occurring in the Telteca 406 SHORT COMMUNICATIONS

I I 2.4 3 WfW _* ’ 0 B 2.0 - -i-r6 GWT /= ’ P .._

CE - - SCC - PGD - CDF

0.4 0.6 0.6 1.0 1.2 1.4 1.6 1.0 Log (x+1) mist-nets captures

FIGURE 3. Relationship between abundance and frequency of sap feeding by bird species during the dry season (simple regression, R* = 0.08, n = 20, P = 0.1 1). Dotted lines give 99% confidence interval. CDF = Common Diuca-Finch, CE = Chaco Earthcrceper, GWT = Greater Wagtail-Tyrant, MP = Monk Parakeet, RWF = Ringed Warbling-Finch, SbC = Shot--billed Canastero, PGD = Picui Ground-Dove, TTS = Tufted Tit- Spinetail, WfW = White-fronted Woodpecker.

Reserve feed to some degree on the fruit of Lycium or the Acorn Woodpecker and other avian species are spp. (Solanaceae), the only fleshy fruit available in the typically regarded as commensal relationships (Miller dry season. and Nero 1983, Kattan and Murcia 1985, Wiens 1989). Phloem sap is both a source of water and a food These woodpecker species do not defend their sap- item with high energy content (Foster and Tate 1966). supplying holes against intruders (but see Foster and Although White-fronted Woodpeckers drill holes Tate 1966, MacRoberts and MacRoberts 1976). In the throughout the year, sap consumption is much greater cast of interactions between White-fronted Woodpeck- during the dry season. Other bird species do not appear ers and facultatively sap-feeding heterospccifics, the to USC this resource at all during the rainy season. relationship is more complex. White-fronted Wood- These other bird species that feed at woodpecker peckers in Telteca do actively defend their sap-flowing drilled holes belong to diverse ecological and taxo- drillings by displacing smaller birds that approach nomic groups. These include both birds that forage pri- them. marily on the ground and those that feed in trees and Aggressive interactions among sap-eating birds sug- shrubs, and both insectivores and granivores (Marone gest the existence of interference competition. Accord- 1992, Marone et al. 1997). The importance of sap as ing to Wiens (1989), when agonistic behavior is related a food resource for these birds in the dry season be- to spatially-defined territories, aggression provides comes evident when comparing the frequency of sap strong evidence of competition. White-fronted Wood- consumption with the total number of feeding events. pcckers live in sedentary groups, as revealed by fre- Such an approach reveals that sap is particularly im- qucnt recaptures and sightings of marked birds on sam- portant to White-fronted Woodpeckers and Monk Par- pled sites. Furthermore, they defend their territories akeets (Myiopsitta monachus). against conspecific intruders. Trees with sap-holes Although some of the most abundant species in the “belong” to specific groups within the different terr- Telteca Reserve do not eat sap, it is clearly an impor- tories. To assert that interactions arc competitive re- tant resource for certain rarer species including Monk quires that sap be a limiting resource. Several obser- Parakeet and Golden-billed Saltator (Saltutor aurantii- vations suggest that this is true. But for a few excep- rostris). In the semi-arid Chaco, only the Greater Wag- tions, sap only is accessible to heterospecific birds at tail-Tyrant showed a close association with woodpeck- woodpeckers’ drillings. Only some Prosopis flexuosa ers’ drillings (Genise et al. 1993), although this con- trees arc used for the purpose of sap consumption, and clusion is probably largely a reflection of this species’ intensity of this use varies among trees based upon the abundance in the arca. In the Telteca Reserve, I also presence of wounds in branches and trunks. However, found Greater Wagtail-Tyrant to bc one of the most the use of such trees is temporally dynamic. Trees may frequent visitors at White-fronted Woodpeckers’ holes. be used in successive years, but not continuously (pers. However, in comparing the rate of sap eating with oth- observ.). For example, in the present study, the use of er foraging activities, it appears that other less abun- sap holes in five of six studied focal trees of P. je,r- dant bird species may be more dependent upon sap uosa notably diminished or ceased over the course of than wagtail-tyrants. one month’s observations. In the semi-arid Chaco, Associations between the Yellow-bellied Sapsucker Genise et al. (1993) found that only 23% of Aspido- SHORT COMMUNICATIONS 407

sperrna quebrucho-blanco (quebracho blanco), the tree taria General de la Organizacidn de 10s Estados species most used by woodpeckers, exhibited wounds. Amcricanos, Washington, DC. Furthermore, many of these trees were re-used in sub- DUNNING, .I. B., AND J. H. BROWN. 1982. Summer rain- sequent years (Gcnisc et al. 1993). Finally, in the Tel- fall and winter sparrow densities: a test of the food teca Reserve, larger wooded patches supported several limitation hypothesis. Auk 99: 123-l 29. groups of White-fronted Woodpeckers, and possibly FOSTER, W. L., AND J. TATE. 1966. The activities and the number of P. &~uosa trees available per group coactions of at sapsucker trees. Living might be fewer than in smaller patches; in support of Bird 5:87-l 13. this contention I observed woodpeckers dwelling in GENISE, J. I?, R. I. STRANECK,AND P HAZELDINE. 1993. larger patches to generally be more aggressive towards Sapsucking in the White-fronted Woodpecker Me- hctcrospccifics at sap holes compared to woodpeckers lunerpes cuctorum. Ornithol. Neotropical 4:77- belonging to single groups occupying smaller patches. 82. Sap is a major food source for White-fronted Wood- HEJL, S. J., J. VERNEK, AND G. W. BELL. 1990. Se- peckers in arid and semi-arid environments in western quential versus initial observations in studies of Argentina, particularly during the dry season. Holes avian foraging. Stud. Avian Biol. 13: 166-173. drilled by woodpeckers enable other species to use the HOLMES, R. T. 1990. Food resource availability and USC sap, facilitating their access to a food item rich in water in forest bird communities: a comparative view and sugar during the season when availability of such and critique, p. 387-393. In A. Keast [ed.], Bio- resources are limited. Given the importance of sap to geography and ecology of forest bird communi- the diet, this food item presumably plays a relevant tics. SPB Academic Publishing, The Hague, The role in both the individual survival and population pcr- Netherlands. sistence of various spccics in this desert environment. KATTAN, G. 1988. Food habits and social organization Interactions among species are likely to have very of Acorn Woodpeckers in Colombia. Condor 90: different consequences for the birds involved. My ob- 100-106. servations rcvcal the cxistencc of an intricate nexus of KATTAN, G., AND C. MURCIA. 1985. Hummingbird as- positive and negative interactions, that can be defined sociation with Acorn Woodpecker sap trees in Co- as a combination of facilitation and competition. Al- lombia. Condor 87:542-543. though the White-fronted Woodpecker displaces many MACROBEKTS, M. H., AND B. R. MACRORERTS. 1976. birds from drillings, these heterospecific avian species Social organization and behavior of the Acorn still gain access to a food resource that would other- Woodpecker in central coastal California. Orni- wise be unavailable to them. Alternatively, observed thol. Monogr. 21. agonistic interactions, especially those between White- MARONE, L. 1992. Seasonal and year-to-year fluctu- fronted Woodpeckers and individuals of smaller bird ations of bird populations and guilds in the Mon- species, strongly suggest that woodpeckers compete te desert, Argentina. J. Field Ornithol. 63:294- for sap. This assertion is strengthened by observations 308. indicating that sap is a limiting resource for birds. Fur- MARONE, L., J. LOPEZ DE CASENAVE, AND V. R. CUETO. ther experimental studies should evaluate the effect of 1997. Patterns of habitat selection by wintering sap availability on the survivorship of individuals and and breeding granivorous birds in the central persistence of populations of the various bird species Monte desert, Argentina. Rev. Chil. Hist. Nat. 70: which rely on this resource. 73-8 1. My thanks to M. Alvarez, R. Ojeda, N. Giannini, R. MILLEK, R. S., AND R. W. NERO. 1983. Hummingbird- Brandl, and the members of the Biodiversity Research sapsucker associations in northcm climates. Can. Group (GIB) in the Institute of Arid Zone Research J. Zool. 61:1540-1546. (IADIZA) who critically read and commented on a RECHER. H. E. AND V. GEBSKI. 1990. Analvsis of the preliminary draft of this manuscript. I also thank G. foraging ecology of eucalypt forest bir&: sequen- Debandi for assisting in diet analyses, M. Cony for his tial versus single-point observations. Stud. Avian useful cooperation, and N. Hor&k and S. Loughccd for Biol. 13:174-180. the English version of the manuscript. This study was SHORT, L. L. 1982. Woodpeckers of the world. Dela- supported by IADIZA and the National Council for ware Mus. Nat. Hist. Monogr. 4. Scientific and Technical Research (CONICET). WIENS, J. A. 1989. The ecology of bird communities. Cambridge Univ. Press, &mbridgc. LITERATURE CITED WIENS. J. A. 199 1. The ecologv of desert birds, R. 278- CABRERA, A. L., AND A. WILLINK. 1980. Biogcografia 310. In G. A. Polis [edy, The ecology of desert dc America Latina. Seric de Biologia 13, Secre- communities. Univ. Arizona Press, Tucson, AZ.