EFFECTS OF SPRING LEAF-FALL ON COMPOSITION AND DENSITY OF BREEDING BIRDS IN TWO SOUTHERN ARIZONA WOODLANDS

RUSSELL P. BALDAl

Department of Zoology University of Illinois Champaign, Illinois 61820

Nesting bird composition and population den- tailed information on these plant associations sities reflect the suitability of a given plant in southern Arizona may be found in Shreve community for providing the requisites for ( 1915), Wallmo ( 1955), Marshall ( 1957), and survival and reproduction, and also, the selec- Lowe (1961,1964). tion of these suitable habitats by the birds. The oaks mentioned above are often re- Two important features in habitat selection ferred to as “live” oaks because they retain are availability of food and nest-sites (Hilden ’ their leaves throughout the winter months. 1965). Open-nesting species depend on foli- Marshall (1957) noted that oak leaves turned age for concealment and protection of the nest brown and/or were dropped during drought from harsh weather and predators (Nice 1957). conditions, and that new leaves appeared More than ample foliage for nest-sites is usu- when more mesic conditions followed. Shreve ally available to these species, whereas a major (1915) mentioned that these oaks shed most, problem confronting hole-nesting species is if not all, leaves in late April or May. A bisea- locating suitable sites (von Haartman 1957). sonal rainfall pattern exists in this area. Heavy Tree-foliage nesting birds of the oak wood- leaf-fall occurs during a spring drought period lands in southeastern Arizona may be limited which follows the winter rainy season. Sum- in the same manner as hole-nesters because mer rains usually begin in late June or early a portion of the tree foliage is absent during July. a period in spring and summer when most Few observations have been made on the birds initiate nesting. birds of these woodlands except by Marshall These woodlands of the desert mountains ( 1957), who analyzed the pine-oak avifauna, can be subdivided into two plant associations: and Balda ( X%9), who studied foliage use evergreen or live oak woodland, and pine-oak by birds. Marshall concluded that both oak woodland. The oak woodland lies on gentle woodland and pine-oak woodland are climax slopes and flat sandy plains below the pine-oak communities, but that the avifauna of the woodland. The lower limit of oak woodland is pine-oak woodland is a unique mixture of bird partly dependent on slope exposure and soil species from the oak woodland below and type but it is generally at about 4509 ft, where ponderosa pine forest above. He did not com- this woodland merges with grassland, forming ment on the effects of spring leaf-fall of the a savanna. The upper limit of the pine-oak oaks on the breeding birds of either woodland. woodland occurs at about 7800 ft, where pon- The pattern of leaf-fall, flowering, and ap- derosa pine (Pinus ponderosa) becomes the pearance of new oak leaves exerts two prin- dominant and predominant tree. These, or cipal effects on the diversity and density of very similar woodlands, extend well into breeding birds in these two woodlands in the Mexico (Leopold 1950). In the Chiricahua Chiricahua Mountains of Arizona. First, the Mountains, Cochise County, Arizona, Emory loss of oak leaves may result in a loss of poten- oak (Quercus emoryi) and Arizona oak (Q. tial nest-sites for birds that nest in tree foliage. a&o&a) predominate. The pines of the Second, the flowering and growth of new higher woodland are Chihuahua pine (Pinus leaves attracts large numbers of insects that serve as an abundant source of food for birds leiophyllu) and Apache pine (Pinus engel- that nest in other sites. mannii), with a scattering of ponderosa pine. The woodlands also contain relatively high METHODS AND MATERIALS densities of various shrubs and succulents. Observations were made in two woodlands of the Alligator juniper (Juniperus ckppeanu) varies Chiricahua Mountains in June 1964 and February- in abundance in the two woodlands. More de- August 1965. Detailed plant sampling in each wood- land was done in conjunction with a bird censusing program to obtain an understanding of some of the 1 Present address: De arhnent of Biological Sciences, North- ern Arizona University, P lagstaff, Arizona 86001. basic plant-bird relationships that exist in these wood- TheCondor, 72:325-331, 1970 r32.51 326 RUSSELL P. BALDA

FIGURE 1. The oak woodland study area showing well scattered Quercus enroryi in full leaf during the month of February.

lands. Trees three or more inches in DBH were sam- RESULTS AND DISCUSSION pled with the point-quarter method (Cottam and Cur- tis 1956). A total of 30 points (120 trees) was sam- OAK WOODLAND pled in both woodlands. Detailed notes were kept This 36-acre study area was within the Chiri- on the growth and flowering of all predominant plants. Birds were censused with the spot-map cahua National Forest and was used for graz- method ( Kendeigh 1944) to obtain an estimate of the ing by the local rancher. The plot was 9 mi. absolute breeding population of each species. Den- SW of Apache, New Mexico and 0.8 mi. NW sities of birds are expressed as numbers of breeding of the Glass Ranch on an unpaved road lead- pairs per 100 acres. Non-territorial species were. cen- sused on the basis of average number seen. The areas ing into Price Canyon. were thoroughly searched for nests, and details such Vegetation. The vegetation was park-like in as nest location, time of egg laying, etc., were re- appearance, with scattered trees of almost uni- corded. Eleven complete daylight counts and four night counts were taken in the oak-juniper-pine wood- form height (fig. 1). There were 68 trees per land study plot, and 10 daylight and three night acre, with Quercus emoyi predominant (table counts were made in the oak woodland. Data pre- 1). The absolute density of oaks was 66 indi- sented herein were analyzed primarily with reference viduals per acre. The shrubby and succulent to the numbers of pairs and species that place their nests in various locations. understory of this area contained predomi-

TABLE 1. Tree species composition, density, and dominance of the two woodlands in the Chiricahua Moun- tains.

Oak woodland Oak-juniper-pine wnodland

R&t~$ Relative ye;; Relative SDecies dominance ’ dominance

Quercus emolyi 95.0 96.5, 20.8 10.7 Quercus arizonica 2.9 2.7 17.4 11.0 Juniperus deppeana 1.4 0.5 33.3 GO.1 Juniperus monosperma 0.7 0.2 - Pinus leiophyllu - - 21.7 lF9 Pinus cembroides - - 1.8 0.3 EFFECTS OF SPRING LEAF-FALL ON BREEDING BIRDS 327 nately the following: Fallugia paradoxa, foliage nesters, two nested in trees and two in Brickellia sp., Rhus microphyllu, Calliandra shrubs and succulents. sp., and Nolina microcarpa. Total shrub and Three important tree-foliage nesters (con- succulent density was 308 individuals per acre. tributing 41 per cent of nests in tree foliage) The phenology of oak leaf-fall, and replace- apparently were able to nest successfully in ment is shown in figure 2. The two species of the area because of their ability to use the oaks which contribute 99 per cent of the foliage dense foliage of Juniperus deppeanu for at (see dominance figures, table 1) in this com- least a portion of their nest-sites. The Chip- munity were in the process of replacing their ping Sparrow had three of five nests in juniper. leaves during most of the spring and early Five of six nests of the Black-throated Gray summer. It is of interest to note that Quercus Warbler were located in juniper. The sixth emoryi did not complete new leaf growth un- nest was built in Quercus emoryi, but the in- til the advent of the summer rains in July. cubating female deserted shortly after the Breeding birds. This plant community sup- leaves fell. Three of seven nests of the Black- ported 36 species of nesting birds, with a total chinned Hummingbird were in juniper. A density of 224 pairs per 100 acres (table 2). female Mexican Jay whose nest was located in The four most abundant species (Bewicks’ an oak also deserted after incubating for three Wren, Ash-throated Flycatcher, Bridled Tit- days, during which time the nest was com- mouse, Lark Sparrow) nested either in natural pletely exposed as leaf-fall was complete. cavities, old woodpecker holes, or on the Thus, some species attempted to nest in oak ground. at the very time leaf-fall was underway, and In most areas where cavity nesting species there were strong indications that many failed. are common, the holes are provided by wood- It is also of interest to note that the number peckers. In this area, however, woodpeckers of singing male Chipping Sparrows on the and old woodpecker holes were noticeably area until the second week in April indicated rare. Most cavities were located at decayed a nesting density of 16 pairs per 100 acres, or places on oaks where large branches had six more than actually nested there. One male broken off the trunk. Some of the oaks sam- Western Wood Pewee and one Gray Vireo pled in this area had hollow trunks because of also appeared to leave established territories heartwood rot. Thus, after the branch stub about mid-April. It is possible that these birds rotted away, potential nest-sites became avail- deserted their territories when oak leaf-fall able but certainly not plentiful or obvious. Of occurred. Thus, some individuals may leave the 24 apparently suitable cavities located in their territories without attempting to nest the study plot and permanently marked and in- once leaf-fall is underway. spected throughout the nesting season, 10 (42 Breeding commenced in this area about the per cent) contained active nests. The total first week in April but some species were still nesting density of cavity nesters was 76 pairs initiating nesting through the first week in per 106 acres and made up almost 34 per cent June (fig. 2). The first species observed to of the total breeding bird population ( table 2). begin nesting were the Black-chinned and Four species nested on the ground, with a Broad-tailed Hummingbirds, which were ob- served gathering nest material from 1 April total nesting density of 24 pairs per 100 acres, on. At this time Quercus emoryi was still in These totals are low for a number of reasons. full leaf. Other early nesters include the Com- The topography of the area varied little, with mon Bushtit, Plain Titmouse, Scotts’ Oriole, few rocky outcroppings, washes, or slopes to and Mexican Jay. add habitat diversity. Also, few fallen logs or branches added debris to the woodland floor. OAK-JUNIPER-PINE WOODLAND The area was heavily grazed by cattle, leaving This study area consisted of 36 acres of un- little standing grass, and new growth was not disturbed woodland located within a relatively forthcoming until the advent of the summer homogeneous tract extending along the south rains in July. There was, therefore, little diver- side of Cave Creek about 0.10 mi. S of the sity of habitat and only limited area suitable Southwestern Research Station of the Ameri- for ground nest-sites. can Museum of Natural History. Vegetation. Juniperus deppeana, a densely All 24 foliage-nesting species had relatively foliated evergreen, is the most abundant tree low densities, totaling 121 pairs per 100 acres. species (table 1). Oaks, with an absolute den- Of these 24 species, 10 nested in the shrubs sity of 46 individuals per acre, contributed and succulents and contributed 36 per cent of only 22 per cent of the foliage present in this the foliage nests. Of the four most abundant area, (as compared with 99 per cent in oak 328 RUSSELL P. BALDA

TABLE 2. Densities of breeding birds and nest-sites in two Arizona woodlands.

Oak Oak-juni er- woodland pine wooB land Nest Species site* pairs per 100 acres

Foliage nesters

Black-throated Gray Warbler (Dendroica nigrescens) T 14 24 Mockingbird ( Mimus voluelottos1 S 12 Chipping Sparrow (S&e% pass&inu) T 10 ii Scotts’ Oriole ( Icterus parisorum) S 10 2 Black-chinned Hummingbird ( Archilochusalexrmdri) T 21 Mexican Jay ( Aphelocoma ultramarina) i 6 Common Bushtit (Psaltriparus minimus) ; 8 10 Gray Vireo (Vireo &it&r) T 8 - House Finch (Carpodacus mexicanus) S 8 Black-headed Grosbeak (Pheucticus melanocephulus) T 6 s Brown-headed Cowbird (Molothrus ater) T, S 4 6 Western Kingbird (Tyrannus verticalis) S 3 - Cassins’ Kingbird ( Tyrannus uociferans) 3 Broad-tailed Hummingbird (Selasphorusplatycercus) ; 3 s Western Wood Pewee (Contopus sordidulus) T 3 5 Bells’ Vireo (Vireo be&i) S 3 - Rufous-sided Towhee (Pip& erythrophthalmus) S 3 - Mourning Dove (Zenaidura mucroura) S 1 Rivolis’ Hummingbird (Eugenes fulgens) T 1 s Scrub Jay ( Aphelocomucoerulescens) S 1 - Hooded Oriole (Zcterrrscucullutus) T 1 Hepatic Tanager (Piranga flaoa) T 1 -i Crissal Thrasher ( Toxostomadorsale) S 1 - Brown Towhee (Pip80 fuscus) 1 Robin (Turdus migratorius) + - ? Solitary Vireo (Vireo solitarius) T - 5 Blue-throated Hummingbird (Lampornis clemenciae) T - 3 Blue-gray Gnatcatcher (P&opt& caerdea) S - Graces’ Warbler ( Dendroiea graciae) T - 33 Buff-breasted Flycatcher (Empidonax fulvifrons) T - 3 Huttons’ Vireo ( Vireo huttoni) T - 2

Total foliage nesting species and pairsb (24) 121 (19) 151

Tree nesters (14) 78 (17) 146

Shrub and succulent nesters (10) 43 ( 2) 5

Cavity nesters Bewicks’ Wren ( Thyomanes bewickii) 27 9 Ash-throated Flvcatcher t Mviarchus cinera-scens) 18 12 Bridled Titmouse (Parus hdlweberi)‘ 15 18 Plain Titmouse (Parus inornatus) 13 10 Screech Owl (Otus ado) 3 White-breasted Nuthatch ( Sitta carolinensis) - ? Eastern Bluebird (S&a ~&is) - 7 Olivaceous Flycatcher (Myiarchus tuberculifer) - 3 Elf Owl (Micrathene whitneyi) - 3 Pygmy Nuthatch ( Sitta pygmaea) - 2

Total cavity nesting species and pairsb (5) 76 ( 9) 71

Ground Nesters Lark Sparrow (Chondestesgrammacus) 15 - Rufous-crowned Sparrow ( Aimcphila ruficeps) 6 11 Harlequin Quail (Cyrtonyx montezumae) 2 - Poor-will t Phalaetwvtilus nuttallii ) 1 1 Painted Redstart ( S&phuga pi&a)’ - Whip-poor-will (Caprimulgus uociferus) - t Turkey (Meleagris gallopauo) - 3

Total ground nesting species and pairsb (4) 24 (5) 25 EFFECTS OF SPRING LEAF-FALL ON BREEDING BIRDS 329

TABLE 2. Continued

Oak Oak-juni er- woodland pinewoo diand Nest Species site’ pairs per 100 acres

Cavity Excavators Red-shafted Flicker (Colaptes cufer) 1 8 Acorn Woodpecker (Meherpes formiciuorus) 1 Arizona Woodpecker (Dendrocopos arizonae) 1 :

Total cavity excavating species and pair? ( 3) 3 (3) 29

Grand totals (36) 224 (36) 267 8 T = tree nestersi S = shrub and succulentnesters. b Number of speciesin parentheses.

woodland). They also dropped their leaves density of 267 pairs per 100 acres (table 2). in the spring (fig. 2). The shrub and succu- The three most abundant species, comprising lent understory contained 13 species contrib- 28 per cent of the total breeding pairs, were uting a total of 478 individuals per acre. The tree foliage nesters. Woodpeckers and, con- most abundant species were in the genera sequently, old woodpecker holes were com- Mimosa, Chrysothamnus, Opuntiu, and Agave. mon in the live and dead pines found in the This study area is described in greater detail area. Heartwood rot and natural cavities were elsewhere ( Balda 1969). not common. A total of 32 potential nest-site Breeding birds. This plant community sup- cavities were located, but only six (19 per ported 36 species of nesting birds with a total cent) contained active nests. Cavity nesting pairs made up 27 per cent of the total nesting pairs. The five species nesting on the ground had OAK-JUNIPER-PINE a total nesting density of 25 pairs per 100 acres I WOODLAND (table 2). This area, which was not grazed, provided a good cover of tall grass and had a relatively rugged topography containing nu- merous rocky outcroppings, dry washes, and scattered boulders. These features, plus dead branches and fallen trees, contributed to the diversity of the woodland floor, and provided diverse and ample nest-sites. A total of 19 species nested in the foliage, of which 17 nested in trees and only two in shrubs and succulents. The latter contributed only three per cent of the foliage nests. LE OAK WOODLAND I Breeding was initiated in the first week of April, when the Bewicks’ Wren began nesting. During the second week the Common Bushtit, Broad-tailed and Black-chinned Humming- birds commenced nesting. Initiation of nest- ing reached a peak in the third week of May, when oak leaves were missing or very small, but most birds nested in either juniper or pine.

CONCLUSIONS AND SUMMARY The sequence of leaf-fall by the two species of oaks followed the same pattern. Quercus

FIGURE 2. Leaf-fall and nesting in two woodlands. arinonica dropped its leaves two weeks before Black portions of horizontal bars indicate full leaf, Q. moyi on both plots. Leaf-fall started three hatched sections indicate leaf-fall and replacement, weeks earlier in the oak woodland than in open sections indicate trees bare. Oaks contributed the oak-juniper-pine woodland. During the 99 per cent of all tree foliage in the oak woodland time of leaf-fall and new growth, almost all and 22 per cent in the oak-juniper-pine woodland. Histograms indicate per cent of species initiating nest- species of birds in both areas began nesting. ing by weekly periods. Since oaks predominate in the oak woodland 330 RUSSELLp. BALDA but not in the oak-juniper-pine woodland, the woodland, even though more holes were pres- 10~sof oak foliage should be of greater conse- ent in the former area. Each of the five spe- quence to the foliage nesting birds of the cies averaged 15.2 pairs per 100 acres, whereas former area. the oak-jumper-pine cavity nesters averaged Comparison of nest-site utilization in the only 7.9 pairs per 100 acres. Three of the four two woodlands yields some striking differ- species of cavity nesters (excluding wood- ences. In the oak woodland plot there were 24 peckers) found in both study areas had higher foliage-nesting species with a total density of populations in the oak woodland and contrib- 121 pairs per 100 acres, whereas 19 foliage- uted 27 more pairs there. nesting species with 151 pairs per 100 acres In the oak woodland 19 species used nest- nested in the oak-juniper-pine study area. Of sites other than tree foliage, as did 16 species the 24 oak woodland foliage nesters, 12 also in the oak-juniper-pine woodland. The 19 oak occurred in the oak-juniper-pine woodland. woodland species had 42 pairs more per 100 The other 12 show a stronger preference, acres than their counterparts in the oak-juni- based on densities elsewhere, for either desert per-pine woodland. or riparian habitats. Of the 12 species common The number of pairs expected to nest in to both areas, nine had lower densities in the each type of site was calculated for the oak oak woodland. The Wilcoxon matched-pairs woodland by using the oak-juniper-pine wood- signed-ranks test indicates this difference as land percentages. On the basis of Chi-square, significant at the 0.025 level. Among these the utilization of the various types was signifi- nine species there were 59 fewer pairs per 100 cantly different for the two areas. acres in the oak woodland. The biggest dif- It appears that the lower density of tree- ference in density between the two areas was foliage nesters in the oak woodland is partially found for the Chipping Sparrow, Black- offset by higher densities of birds nesting in throated Gray Warbler, and Black-chinned sites other than tree foliage. This is supported Hummingbird, all of which are known to use by the fact that a much higher percentage of both oak and jumper for nest-sites. The low cavities was occupied in the oak woodland; density of juniper, combined with the loss of also, more species with higher densities nested oak foliage, apparently restricted their abun- in shrubs and succulents. dance in the oak woodland (cf. Balda 1969). The almost total absence of tree foliage in Tree-foliage nesters averaged 5.6 pairs per the oak woodland during the start of the nest- species in the oak woodland and 8.6 pairs per ing season may be an important limiting factor species in the oak-juniper-pine woodland. for species which normally place their nests Five of the seven tree-foliage nesters found in tree foliage. However, their resulting low exclusively in the oak-juniper-pine woodland densities may reduce demands for food, so are known to select conifers for nest-sites and that species which can procure nest-sites, such would therefore avoid pure stands of oaks. as cavity nesters or species which nest in These five species (Robin, Solitary Vireo, shrubs, succulents, evergreens or on the Graces’ Warbler, Blue-throated Hummingbird, ground, may be able to procure ample food to and Buff-breasted Flycatcher) all had rela- maintain relatively higher populations. The tively low densities in oak-jumper-pine wood- oaks of both woodlands, when in flower, at- land. tracted large numbers of insects and conse- Ten species in the oak woodland nested in quently large numbers of foraging birds, but shrubs and succulents where they were unaf- of course provided little foliage for nest con- fected by oak leaf-fall. These 10 species av- cealment. In the oak-juniper-pine woodland, eraged 4.3 pairs per 100 acres, whereas only oak flowers also attracted insects (Balda 1969), two species, with an average of 2.5 pairs per but there were considerably fewer oaks pres- 100 acres, selected nest-sites in the understory ent and ample tree-foliage nest-sites were in the oak-juniper-pine woodland. The low available in the pines and juniper. Von Haart- densities of species nesting in shrubs and suc- man (1957) emphasizes that the benefits de- culents in the latter area do not reflect a lack rived from nesting in tree cavities may be of nest-sites, as there were more shrubs and partly offset by severe competition resulting succulents present in the oak-juniper-pine from lack of suitable holes. The present study community. suggests that in the oak woodland, a unique Although there were four more species of and geographically limited plant association, cavity nesters present in the oak-juniper-pine a lack of suitable tree foliage sites may limit woodland, their total nesting density was less the density of tree-foliage nesting birds. than that of the cavity nesters in the oak It appears that ample insect food, associated EFFECTS OF SPRING LEAF-FALL ON BREEDING BIRDS 331 with the flowers and new oak leaves, is avail- and criticism during all phases of the study. I am able to satisfy reproductive demands. The particularly grateful to the Frank M. Chapman Memo- rial Fund of the American Museum of Natural History usual consumers, tree foliage nesters, however, and the University of Illinois Research Board for are limited in density by a factor other than financial support. interspecific competition with birds that nest elsewhere, i.e., the untimely loss of tree fo- LITERATURE CITED liage. BALDA, R. P. 1969. Foliage use by birds of the The question arises as to whether the dif- oak-juniper woodland and ponderosa pine forest ference in density of tree foliage nesters is a in southeastern Arizona. Condor 71: 399412. COTTAM, G., AND J. T. CURTIS. 1956. The use of result of increased predation, or simply a fail- distance measures in phytosociological sampling. ure of some individuals to establish and main- Ecology 37:451460. tain territories in the oak woodland. My mea- HILDEN, 0; 1965. Habitat selection in birds. A ger data show no difference in intensity of review. Ann. Zool. Fenn. 2:53-75. KENDEIGH, S. C. 1944. Measurement of bird popu- nest predation between woodlands, nor be- lations. Ecol. Monogr. 14:67-H%. tween different nest-sites within either wood- LEOPOLD, A. S. 1950. Vegetation zones of Mexico. land. Ecology 31:507-518. The lower densities of species nesting in LOWE, C. H., JR. 1961. Biotic communities in the Sub-Mogollon region of the inland southwest. J. tree foliage in the oak woodland may be due, Arizona Acad. Sci. 2:40-49. in part, to a lack of initial selection of territo- LOWE, C. H., JR. 1964. Arizona landscapes and ries and/or a desertion of territories or nests habitats. p. 1-132. In C. H. Lowe, Jr. [ed.] when an important sign stimulus, foliage, is The vertebrates of Arizona. Univ. Arizona Press, Tucson. removed. Individual nests started in trees be- MARSHALL, J. T., JR. 1957. Birds of pine-oak wood- fore leaf-fall may be deserted because of lack land in southern Arizona and adjacent Mexico. of concealment and protection, or because the Pacific Coast Avifauna no. 32. microhabitats surrounding these nests have NICE, M. M. 1957. Nesting success in altricial birds. Auk 74:305-321. been seriously altered by the loss of foliage. SHREVE, F. 1915. The vegetation of a desert moun- One important factor in this regard may be tain range as conditioned by climatic factors. an increase in solar radiation. Carnegie-Inst. Wash. Publ. no. 217. VON HAARTMAN. L. 1957. AdaDtation in hole-nest- ACKNOWLEDGMENTS ing birds. kvolution 11: 33e47. WALLMO, 0. C. 1955. Vegetation of the Huachuca I wish to thank the following for critically reading the Mountains, Arizona. Amer. Midland Nat. 54: manuscript: T. Vaughan, C. D. Johnson, J. Rominger, 466-460. and R. Hevly. J. D. Ligon was instrumental in initiat- ing this analysis, while S. C. Kendeigh offered advice Accepted for publication 30 December 1969.