The Condor92:866-8885 0 The CooperOrnithological Society 1990

BIRD DISTRIBUTION PATTERNS IN RIPARIAN HABITATS IN SOUTHEASTERN ’

THOMAS R. STRONG* AND CARL E. BOCK Department of Environmental, Population, and OrganismicBiology, Universityof , Boulder, CO 80309

Abstract. speciesdensities were determined for summerand winter on 132 study plotsgrouped into 25 riparian habitatsin or near the HuachucaMountains of southeastern Arizona. The habitatswere defined based on the dominantriparian tree species, the sizeof the riparian stand,and the type of adjacentupland vegetation.Vegetation characteristics and physicalenvironmental data were collected at eachplot. The typeof dominantriparian tree speciesinfluenced bird speciesrichness and total densityduring the breedingseason. Cottonwoodhabitats had the greatestrichness, and bothcottonwood and sycamorehabitats had high densities.Upland vegetationwas an important factorrelated to winter species richnessand abundance,with plots in open grasslandareas having greaterrichness and density.Riparian standsize was a relativelypoor predictor of avian densityor richnessin eitherseason. Groups of bird speciesthat sharedsimilar density distributions in the summer wereassociated with specificriparian habitats. The winter patternof speciesgroups was not as clear,and groupscould not he assignedto riparian habitats,but they were relatedto eitherwooded or open upland vegetation.Riparian habitatswere also clusteredbased on similar densitiesof . In summer,high-elevation habitats were distinctfrom low-ele- vation and foothillhabitats. In winter,riparian habitats separated into categoriesof wooded vs. openadjacent vegetation. Key words: Huachuca Mountains: Arizona: riparian; speciesrichness; species densities; avian communities.

INTRODUCTION terminants of the abundance and variety of birds Riparian woodlands comprise less than 0.5% of associatedwith a particular stand of riparian veg- the land area of Arizona, yet they support an etation: (1) dominant tree species(Pase and Lay- extraordinary variety and abundance of birds ser 1977); (2) patch size (e.g., Robbins 1979, Sza- (Johnson et al. 1977, Szaro 1980). The bird life ro and Jakle 1985, Blake 1986); and (3) nature of cottonwood (Pop&s fremontii), mesquite of adjacent (upland) vegetation (Stevens et al. (Prosopis spp.), and exotic saltcedar (Tamarix 1977). We measured the vegetation and physical chinensis) woodlands in lowland river valleys has environmental characteristics of 132 small ri- been well studied (e.g., Carothers et al. 1974, parian plots, and we counted their birds for three Anderson and Ohmart 1977, Stamp 1978, Szaro summers and two winters between 1984 and and Jakle 1985). Riparian woodlands at higher 1986. Each plot was dominated by one of seven elevations in the Southwestinclude a greater va- riparian tree species,belonged to a large (greater riety of trees and shrubs than the river valleys than 1,000 m) or small (less than 200 m) linear (Pase and Iayser 1977), but their bird popula- riparian patch, and was adjacent either to open tions have not been as thoroughly examined mesquite grasslandor to woodlands of and/ (Szaro 1980). The purpose of the present study or . was to describe patterns of bird speciesdistri- The data collected during this study were an- bution and abundance in riparian woodlands of alyzed with the goal of determining (1) if certain the Huachuca Mountains and vicinity in south- riparian habitat types supported significantly eastern Arizona. richer or more abundant avifaunas, (2) if the bird At least three factors could be important de- species grouped themselves into discrete, rec- ognizable assemblages,(3) if these assemblages were associatedwith predictable vegetation and/ ’ ’ Received15 December1989. Final acceptance22 May 1990. or physical environmental patterns, and (4) if 2Present address: 3260 Walnut Ave. SW, Seattle, such patterns and assemblagesdiffered between WA 98116. summer and winter. DISTRIBUTION IN RIPARIAN HABITATS 861

MATERIALS AND METHODS range. These included O’Donnell and Post can- yons on the National Audubon Society Apple- STUDY AREA ton-Whittell ResearchRanch and the Canelo Hills This study was conducted in the vicinity of the Preserve owned by the Nature Conservancy; Huachuca Mountains in Cochise and Santa Cruz Huachuca, Garden, Sawmill, Scheelite, and counties in southeasternArizona. The Huachuca Blacktail canyons on the Fort Huachuca Military Mountains reach a maximum elevation of 2,885 Reservation; the Babocomari River and Vaughan m (9,466 ft) at Miller Peak, while the surrounding Canyon on the Babocomari Ranch; and Lyle and valleys are about 1,350 m. The San Pedro and Woodyard canyons on Coronado National For- San Rafael valleys and the Sonoita Plain are est. semidesert grasslands, grading into oak wood- lands in the foothills. The higher parts of the CENSUS TECHNIQUE range are covered by pine-oak or mixed conif- The variable circular-plot method (VCPM) erous forests.The Huachucas are incised by sev- (Reynolds et al. 1980) was selectedfor this proj- eral deep canyons in a roughly radial pattern. ect. By their nature, the small stands of this pro- Many of these canyons have perennial streams ject could not be censusedby transect methods. or reliable seasonal stream flow. Virtually all VCPM is appropriate for sampling the sites in drainages are subject to intermittent flooding winter as well as in summer. Szaro and Jakle during the winter rains or the late summer mon- (1982) have compared VCPM and the spot-map- soons. The water available in the canyons sup- ping method (SMM) in desert riparian and scrub ports a variety of riparian habitats that in turn habitats near Superior, Arizona. For all but one support diverse bird communities which formed species, the density estimates from SMM were the focus of this project. within the 95% confidence limits of the density estimates from VCPM, and they recommended STUDY SITES VCPM for censusing in small habitat “islands.” Habitats dominated by seven riparian tree spe- Vemer and Ritter (198 5) compared VCPM with cies were included in this project. These tree spe- transect counts in oak-pine woodlands of Cali- cies are the velvet ash (Fruxinus velutina), Fre- fornia. They preferred point counts to transects, mont cottonwood (Populus fremontii), desert but they cautioned that small sample sizespermit willow (Chilopsislinearis), big-tooth maple (Acer density estimates for only a small portion of the grandidentatum), Arizona sycamore (Platanus speciesdetected. They also questioned whether wrightii), Arizona walnut (Juglans major), and the density estimates from either method were willow (Sulix spp.). For each tree species,both acceptably accurate. For the purposes of this large and small standswere selected.For the pur- project, the absolute accuracy of density esti- posesof this project, a large stand was 1,000 m mates was less important than the relative den- or more of continuous riparian habitat along a sity estimatesbetween different plots or habitats. drainage, and a small stand was 200 m or less of Because we used the same census procedure at continuous riparian habitat. The riparian stands each study plot and during each season of data were usually less than 50 m in width, and some collection, we assumedthat the density estimates were less than 20 m. For both large and small for all plots or all habitats were comparable, and stands, areas were selected with either open or that statistical comparisons based on these es- wooded adjacent uplands. An open upland area timates were legitimate. was primarily grassland,but many sitesincluded In applying the variable circular-plot method some mesquite or other desert shrubs.A wooded to this project, we used 5-m increments of radius upland was dominated by (Quercus spp.) out to 40 m from the central point, and 10-m and/or (Pinus spp.). Control sites were se- intervals beyond that out to 200 m. Each plot lected in wooded and open areas along drainages was censusedfor a 5-min period six times each with no riparian trees. Several replicate study seasonduring the breeding seasonsof 1984,1985, plots were selectedfor each combination of fac- and 1986 (May through July), six times during tors. A total of 132 study plots grouped into 25 the winter of 1985-1986, and five times during habitat types (seeAppendix A) were included in the winter of 1984-1985 (December through this project. The study sites were located along February). Becausethe activity level of birds de- several major drainagesnear the north end of the creaseddrastically during the middle part of the 868 THOMAS R. STRONG ANDCARL E. BOCK day, the censustimes were restricted to the first was recorded.Forty points were recordedon each 4 hr after sunrise. To obtain more reliable esti- of the four subplots, with the results combined mates of detection limits, all replicate plots were and converted to percentages. lumped for each riparian habitat type, and the Upland tree abundance was measuredat points resulting detection limits were then used for each 50 m from the plot center on each side of the plot within that habitat. If no individuals of a stream channel. At each point the distance to the specieswere seenwithin that limit at a particular three nearest trees was measured. The reciprocal plot, but they were observed farther away, then of the average of these distances provided an the greater distance was used to calculate the estimate of upland tree density. Canopy coverage species’ density at that plot. at each site was measured with a spherical den- siometer. For the riparian canopy, readings were VEGETATION ANALYSIS taken at the center point and from the center of Vegetation and physical environmental data were each subplot looking toward the center point. For collected at each of the 132 study plots. The the upland canopy, two readings were taken at methods used roughly follow the techniques pro- each upland tree point, one looking upstream posed by James and Shugart (1970) with mod- and one downstream. The canopy height at the ifications as appropriate for these riparian hab- center point was determined by measuring the itats. An area within a radius of 35 m from the angle up to the canopy from a point 35 m from center point of each plot was selectedfor analysis. the center point and using appropriate trigono- This distance was selected based on an average metric relationships. Because hours of sunlight bird speciesdetection limit of 37.3 m reported might be an important factor in the vegetation by Szaro and Jakle (1982). Within this circle, all and bird densities, angles to the east and west riparian trees were identified and measured. Be- horizons were measured from the center point. cause of the tendency of willows and desert wil- The elevation at each site was estimated from lows to branch very close to the ground, the di- USGS topographic maps. ameters of all trees were taken at boot height. In Several factors were used to describe each site heavily wooded areas, the nonriparian trees, pri- qualitatively. These factors included stand size, marily oaks and pines, were counted on four sub- distance to the next stand, presence of water, plots of 10 m radius each. Two plots were cen- upland vegetation, and upland tree density. Each tered at distancesof 15 m from the center point, of these factors was broken down into several and the other two were at distances of 20 m and ordinal categories, which allowed a better de- 25 m. The subplots were oriented such that they scription of the site characteristicsthan the broad did not overlap the drainage, if possible. The groups of large vs. small stands and wooded or subplots represented nearly one-third of the full open adjacent upland. plot. In more open areas, all nonriparian trees were counted. No distinction was made between the speciesof oak or the speciesof pine. From ANALYTICAL PROCEDURES the data collected in the field, densities and basal Bird speciesdensities were calculated for all hab- areasfor all tree specieswere calculated. All trees itats and for each plot within each habitat for were grouped into size classesas follows: less both summer and winter. The number of bird than 30 cm diameter, between 30 and 60 cm, speciesobserved (speciesrichness) and the total and greater than 60 cm. density of birds recorded in each habitat or plot Understory plants and ground cover were also were determined. Differences in total bird measured on 5-m radius subplots with the same population densities and species richness be- centersas the 10-m subplots. Two tape measures tween plots with different riparian tree types, dif- were laid out on the ground, one oriented parallel ferent stand sizes, and different upland vegeta- to the stream channel and the other perpendic- tion were tested with analysis of variance ular, such that they crossedat the center of the (ANOVA) procedures.Comparisons were made subplot. At 0.5-m intervals along each tape, the using pooled data from the three breeding sea- ground cover was recorded as either grass/herb sons or the two winter seasons. The replicate cover or bare ground/litter. At each point, the plots within each habitat type were used to gen- presenceand speciesof shrub or sapling canopy erate the means and variances needed for the DISTRIBUTION IN RIPARIAN HABITATS 869 comparisons. To isolate the effectsof stand size RESULTS or upland vegetation, one-way ANOVAS tested RELATIONSHIPS WIT-J RIPARIAN for one variable while holding the other constant. TREE SPECIES For example, the large wooded standswere com- Summer bird speciesrichness, sorted by domi- pared with the small wooded stands to test for nant riparian tree species(Fig. 1, top), separated stand size effects, or the small wooded stands into four overlapping subsets(F = 16.079, P < were compared with the small open stands to 0.0001). The data used in this figure and others test for effectsof adjacent upland vegetation.Two- are available in Strong (1987). Maple plots, with way ANOVA was not used because of missing low bird speciesrichness, and cottonwood plots, cells in the control, maple, and willow habitats. with high bird species richness, were the only Throughout this paper, the terms “wooded” and groupsthat fell into only one subset.The species “open” will refer to the nature of the vegetation richness of winter birds revealed three overlap- adjacent to the riparian zone. Control plots could ping subsets (F = 8.302, P -c 0.0001). Maple not be assignedto a large or small category and plots, with low speciesrichness, and cottonwood were not included in the analysesbased on stand plots, with high richness, were again the only size. Bird speciesrichness and the total density groups falling into a single subset. for both summer and winter were analyzed with In summer, avian densities in the various ri- the Komolgorov-Smimov goodness-of-fittest for parian habitats separatedinto three overlapping normality. Raw data for speciesrichness could subsets(Fig. 1, bottom; F= 25.532, P < 0.0001). not be distinguished from a normal distribution Maple and control (no riparian trees) habitats for either summer or winter. Raw data for total had low bird densities, while sycamore and cot- density were not normally distributed, but nat- tonwood habitats had much higher densities.The ural log transformations of the density data re- sequenceof riparian tree speciesfrom minimum sulted in distributions that could not be distin- to maximum bird density was similar to that guishedfrom normal for both summer and winter. based on bird species’ richness, but one notable For this reason, natural log transformations of exception was the group of three willow plots. density data were used in all analysis of variance These plots had high densities of relatively few procedures. In each analysis, the SchellZ a pos- species, including the Red-winged Blackbird, teriori procedure was used to identify homoge- Common Yellowthroat, and the Brown-headed neous subsets. Cowbird (see Appendices B and C for scientific Multivariate statistical procedures were ap- names of all bird species).This habitat was unique plied to the data set with the goals of (1) recog- in its high density of very small diameter willows, nizing groups of bird specieswith similar habitat surroundedby marshy grassland.Blackbirds were associations and (2) grouping riparian habitat common breeding birds in the marshy area, and types supporting similar avian assemblages.For yellowthroats were common in the willows and the first objective, Pearson’s product moment in grassof the marsh. correlation coefficients were computed between Total densities of winter birds in winter sep- densities of each bird speciesand the 47 physical arated the riparian tree speciesinto two overlap- and vegetation characteristicsmeasured for each ping subsets(F= 5.623, P < 0.0001). Maple and plot (see Appendix D). These correlations then control habitats again had low bird densities, and were used as ecological descriptors of each bird cottonwood and sycamore habitats had higher species, and the birds were clustered based on densities. The sequencefrom minimum to max- those descriptors. The clustering method used imum bird densities was nearly identical to that was single-linkage (UPGMA, Rohlf et al. 1972). observed in summer, but the range of differences For the secondobjective, the plots were grouped was much lower in the winter data. a priori into major riparian habitat types, based on shared characteristics of dominant riparian RELATIONSHIPS WITH RIPARIAN tree species,stand size, and type of adjacent up- STAND SIZE land vegetation. Mean densities of each bird spe- The first set of comparisons to test the effectsof cies were computed for each habitat type, and stand size was restricted to plots with wooded these densities were used as descriptorsfor clus- uplands (Fig. 2). In summer, control plots had tering the habitat types. lower average speciesrichness than plots in large, 870 THOMAS R. STRONG AND CARL E. BOCK

m SUMMER 0 WINTER 4Or D

(kD CD W T 1, 3 30- T

%! A z -r F ff 20- b,c 1 G i? cn a lo- 1

O-

250- Lu 9 L ii! 200- 0

s a,b

2 2 150- ah 6 B ab T ab ii T / ’ B,C -; -z A,B g 10Ot A A,B cl, a s!l Aa n < 50

_n I 0 CGNTRZL ASH COTTON- CESERT MAPLE SYCAMOf4E WALNUT WI-LOW 11WXQ WILLOW 11 Number of plots: 17 19 23 18 10 22 20 3 FIGURE 1. Means and standard errors of bird speciesrichness and total bird density on plots dominated by various riparian trees in southeasternArizona. For both variables, homogeneoussubsets were distinguishedby the Scheffe a posteriori test with ANOVA. Upper and lower case letters indicate groups that did not differ significantly in summer and winter, respectively. DISTRIBUTION IN RIPARIAN HABITATS 871

m SUMMER II WINTER

BA -i-

/

125 A A 10 T 100 a X 75 T 1

50 X

25

~ ” LARGE SMALL CONTR c IARG il_lCONTROL Number of Plots: 37 25 11 32 21 6 WOODED UPLAND OPEN UPLAND

FIGURE 2. Means and standard errors of speciesrichness and total bird density on plots in large riparian stands, small riparian stands,and control standswithout riparian trees. The comparisonsare restricted to sites locatedin either wooded areasor open grasslands.For eachcomparison, homogeneous subsets were distinguished by the Scheffe a posteriori test with ANOVA. Upper and lower case letters indicate groups that did not differ significantly in summer and winter, respectively.

DISTRIBUTION IN RIPARIAN HABITATS 813

CLUSTERS OF BREEDING BIRD SPECIES Correlation Coefficient I I I I I I Plot densities of 87 breeding bird species(Ap- -0.4 -02 0 0.2 0.4 0.6 pendix B) were correlated with 47 plot vegetation and physical environmental characteristics(Ap- 1,7,10,12,13,23,24. pendix D), and the resulting correlation matrix 30,32,37,50.61.62. was used to generatea dendrogram (Fig. 4). This 68,70.75,86 L 26,57 procedure clustered groups of bird speciesthat shared similar environmental associations. Ap- 6 r-l15,19,34.55,56.75 pendix B also indicates the habitats that were jL 233.16 utilized most frequently by each species. It is 6,8,11,14,31,38.40, apparent from this list that many specieswere 43,47.46,58,74,76. generalistsand could be found in many different 77.76,79,62 habitats, while other species were common in 65,81 only one or two habitats. The first major division in this dendrogram 4,22,35,36.84,87 F was between lowland and foothill species(branch _ 5,9,17,49.60,66 A) vs. highland species(branch E). Within the set of highland speciesgroups, branch I included only the Scott’s Oriole (85; speciesnumbers in this sectionrefer to Appendix B and Fig. 4), which was found in relatively low densities in a wide I 85 range of habitats. The other highland groups in- cluded species associated with specific habitat FIGURE 4. Dendrogram of 87 breeding bird species types. Branch F specieswere found in relatively based on correlations of their densities on 132 study plots with 47 vegetation and physical environmental dry canyons with some maples, and heavily characteristicsat each plot. The speciesnumbers given wooded with oaks. Branch G specieswere most for each branch correspond with speciesnumbers in abundant in wooded canyons with large riparian Appendix B. Cophenetic correlation = 0.874. Major trees and perennial water. Branch H specieswere groups, indicated by letters, are defined in the text. found at the highest elevations in open pine for- est, with a few sycamores. Within the lowland or foothill sets,branch D was somewhat unusual. (3) Common Nighthawk (16), and Turkey Vul- These species, including the American Kestrel ture (2), were widely but sparsely distributed in (4) Acorn Woodpecker (22), Sulphur-bellied lowland and foothill habitats. Flycatcher (36), and Northern Oriole (84), ap- peared to share a strong affinity for large syca- CLUSTERS OF WINTERING BIRD SPECIES more trees, regardless of upland vegetation or A corresponding dendrogram for 60 winter spe- elevation. Within the remaining groupsthere was cies (Appendix C) was based on their correlations a clear and major break between speciesin open with the same physical and vegetation charac- areas with scrubby vegetation (branch C) and teristics (Fig. 5). Appendix C also includes lists speciesin areas with large riparian trees (branch of habitats commonly used by each of the spe- B). The species in branch C were found in a cies. The first major division separated groups variety of habitats, all of which were open, with of speciesmost common in lowland, open areas low stature riparian vegetation. Branch B in- (branch A) from groups of species in wooded cluded species common in lowland areas with areas (branch H). Within the set of open country large riparian trees and perennial water. The species,another important branch isolated birds Northern Flicker (26) and Phainopepla (57) common in open areas with scrubby vegetation shared high densities in a large, wooded cotton- or no trees (branch B). Typical speciesin branch wood stand. The Black-chinned Hummingbird B were the Mourning Dove (9; numbers in this (19), Ash-throated Flycatcher (34), and Northern section refer to Appendix C and Fig. 5), Chi- Mockingbird (55) were found in lowland areas huahuan (23), Loggerhead Shrike (39) with large riparian trees, particularly sycamores, Canyon Towhee (44), and Song Sparrow (49). and in drier areas than those for other speciesin Branch C included the Scaled Quail (5) and branch B. Three species, the Red-tailed Homed Lark (20). The other open area division 874 THOMAS R. STRONG ANDCARL E. BOCK

Correlation Coefficient Correlation coefficient I I I I I I I I I I I -0 2 0 0.2 0.4 0.6 00 0.2 0.4 0.6 0.8 1.0

B 1,7,8,9,10,23,25, 29,31 .32,39,44,46, 47,78,49,50,55 c 5,20

D A c 2,3,17,58,60

E 6,12,18,38,41, 42,45,54

F 15,19,28,40, A 51,57,59 G 52 I 4

13,14,18,21,22, 24,30,36,53 J H 28,35

K ii ,27,33,34, f 6 37,43,58

FIGURE 5. Dendrogramof 60 winter bird species FIGURE 6. Dendrogramof 25 riparianhabitat types basedon correlationsof their densitieson 132 study basedon densitiesof 87 breedingbird specieslisted in plotswith 47 vegetationand physicalenvironmental AppendixB. Wooded= adjacentuplands are oak or characteristicsat eachplot. The speciesnumbers given pine/oak woodlands;open = adjacent uplands are for eachbranch correspond with speciesnumbers in largelygrassland; large = riparian standis greaterthan AppendixC. Copheneticcorrelation = 0.812. Major 1,000m alongstream channel; small = riparian stand groups,indicated by letters,are de6nedin the text. is lessthan 200 m along streamchannel. Cophenetic correlation= 0.767. included bird speciesfound in sites with small speciesin branch K were rather widespread,with clumps of large riparian trees (branch D), areas variable abundances, in wooded areas. These with large treesand mesquite (branch E), or areas species included the Acorn Woodpecker (1 l), with small-stature to medium-stature trees White-breasted Nuthatch (27), Ruby-crowned (branch F). Some representative speciesof these Kinglet (33), and Rufous-sided Towhee (43). groups were the American Kestrel (3), Northern Flicker (17), and Pine Siskin (58) in branch D, CLUSTER OF HABITATS-SUMMER Gambel’s Quail (6), Gila Woodpecker (12), and The first dendrogram of 25 habitats (Fig. 6) is Abert’s Towhee (45) in branch E, and Ladder- based on densities of 87 breeding bird species backed Woodpecker (15), Yellow-rumped War- within those habitats. The clustersof habitats in bler (40), and LesserGoldfinch (59) in branch F. this diagram show a reasonable and biologically Branch G contained only a single species, the meaningful pattern. The first division of the den- Dark-eyed Junco (52), which was widespreadand drogram separatesthe small, open willow habitat abundant in many habitats. (branch B) from all others (branch A). This hab- Within the groups of speciesmore common in itat was unique in being in a marshy grassland wooded areas, branch I included only one spe- area, with a high density of low-stature willows. cies, the Montezuma Quail (4), which was re- The next major division was between high ele- corded in several habitats at relatively low den- vation, wooded habitats (branches C and D) and sities. The species in branch J, including lowland or foothill habitats (branches E and E). Williamson’s Sapsucker (14), Steller’s (2 l), Hermit Thrush (36), and Yellow-eyed Junco (53), CLUSTER OF HABITATS- WINTER were most common in highland canyons that A corresponding dendrogram of the same 25 were heavily wooded with oaks and pines. The habitats using density data for 60 speciesof win- DISTRIBUTION IN RIPARIAN HABITATS 875

tering birds (Fig. 7) revealed a pattern different Correlation Coefficient from that obtained from the summer data. The first division of this dendrogram identified two habitats as having winter bird speciesdensities distinctive from those in all other habitats. These two were the small, open, cottonwood and wil- low habitats, both ofwhich included plots in low, grasslandareas, with a lot of water available. The next branch point was very clean and significant, separating all but one of the riparian habitats Ash,Small,Wwded A F adjacent to open (grassy) uplands (B) from all Co”onwocd,Small,Wocded those adjacent to wooded uplands (A).

DISCUSSION FACTORS INFLUENCING BIRD SPECIES RICHNESS AND TOTAL DENSITY Results of this study demonstrate the importance of riparian tree species,riparian stand size, and adjacent upland vegetation to bird speciesrich- ness and total density in southeastern Arizona. FIGURE 7. Dendrogram of 25 riparian habitat types Speciesrichness and total density varied signif- based on densities of 60 winter bird specieslisted in icantly with the riparian tree species.In general, Appendix C. Definitions are the sameas those in Figure maple habitats had the lowest speciesrichness 6. Cophenetic correlation = 0.866. and density, and sycamore and cottonwood standshad the highestrichness and density. These relationships appeared stronger in summer than phasizing the importance of sycamoresand cot- in winter, although they were highly significant tonwoods to breeding bird species. in both seasons. The generally low bird species richness and The importance of riparian vegetation to bird total density in control habitats were primarily species in arid or semi-arid environments has due to very low numbers of birds in open control been discussedby many authors. In particular, plots, although wooded control plots also were the lower Colorado River valley in Arizona has relatively low. The paucity of birds in maple hab- been the site for much riparian research.Ander- itats was somewhat surprising. These plots were son and Ohmart (1977) correlated bird popula- in narrow, high-walled canyons where they could tion parameters during different seasonsof the receive relatively little sunlight. These conditions year with vegetation parameters along the Col- createda cooler microclimate that may have been orado River. In summer, they found the greatest less attractive to birds than other habitats. Karr numbers of birds in areaswith the greatestamount and Freemark (1983) considered microclimate of total vegetation. However, in fall the greatest to be more important than vegetation factors in numbers were found in relatively open areas,and determining avian speciesassemblages. Petit et in winter, bird densities were most highly cor- al. (1985) found speciesrichness to be positively related with low vegetation (l-3 m). Johnson et correlated with relative humidity in a mature al. (1977) noted that the highest breeding bird deciduous forest in Ohio, but they believed that densities in North America have been reported speciesrichness was directly related to food re- from cottonwood riparian zones in the South- sourcesthat were in turn correlated with relative west, and that loss of these habitats could result humidity. In our study area, the cottonwood hab- in the loss of 47% of breeding bird speciesin this itats were relatively wet and had high species region. Bock and Bock (1984) found that syca- richness, but the large maple habitat also was mores provide both food resourcesand nesting along a permanent stream and had relatively few sites for birds. Sycamores are particularly im- species. portant to the Elegant Trogon (Taylor 1980). Our Other species of riparian trees, such as ash, results concur with these previous studies in em- desert willow, and walnut, are generally smaller 876 THOMAS R. STRONG AND CARL E. BOCK than the cottonwoods and sycamores, and they passerinespecies richness and density in riparian are a lessdominant part of the vegetation. These habitats than in adjacent nonriparian areas in trees tend to be relatively short, and in wooded central and southeasternArizona. Knopf (1985) areas there is not as great a contrast between the also found greater bird species diversity in ri- riparian trees and the adjacent vegetation. Some parian vegetation than in adjacent upland areas of these trees, particularly ash and walnut, tend in an elevational transect in the Platte River to have very hard wood, which might be more drainage in northern Colorado. resistant to decay and have fewer cavities, thus The lack of differencesbetween large and small providing fewer nest sites for several species. stands suggeststhat stand size was unimportant The riparian trees in our study area are all to most bird species,except in open habitats in deciduous, and they are a much less dominant winter. These results would appear to be in con- part of the environment in winter than in sum- flict with the positive species-areacorrelation re- mer. This difference may at least partially ac- ported by Blake (1986) for isolated woodlots, and count for the observed differencesbetween sum- with the speciesdisappearances associated with mer and winter in the patterns of bird species forest fragmentation reported by Whitcomb richness and total density based on riparian tree (1977) and Robbins (1979). However, in a study species.Another important factor is the effect of in small woodlots in the Netherlands, Opdam et migratory species.Many speciesthat are present al. (1985) found no correlation between the num- in this area only as summer breeding speciesare ber of bird speciespresent and variables describ- insectivorous, especially the warblers and fly- ing the isolation of the woodlots. They did find catchers. These speciesappear to be dependent that different speciesshowed different responses on the riparian vegetation, particularly in the to isolation and area of woodlots, and that the open areas. In contrast, many winter migrants speciesmost affectedwere those restrictedto ma- are granivorous speciesthat are not dependent ture woods. Blake and Karr (1984) also reported on riparian zones as a foraging habitat, although bird speciesdifferences in responseto forest size, riparian vegetation doesprovide cover. In a study with long-distance migrants and forest-interior of migratory insectivorous birds in the Chirica- speciesbeing poorly represented in small forest hua Mountains, Hutto (1985) found significant patches.Our results are consistent with different differencesin habitat use between the spring and species showing different preferences based on fall migration seasons,and there were distinct stand size, and migratory species in particular groups of species that shared similar seasonal affectedthe patterns of speciesrichness and den- distribution patterns. He reported that the den- sity in summer and winter. sities of all specieswere significantly positively In our study area, the large riparian habitats correlated with measuresof food availability. In were never part of a continuous, widespread, ho- our study area, it seems very likely that the rel- mogeneoushabitat. They were very narrow strips atively lush vegetation associatedwith riparian restricted to stream drainages, and virtually all zones would provide greater resourcesfor insec- of our study sites could be considered habitat tivorous birds than either surrounding grasslands edges.Even in the widest riparian stand, it usu- or oak forests.Morrison et al. (1986) also found ally was not possible to be more than 50 m from differences in bird species habitat use between upland, nonriparian vegetation. These areasthen summer and winter in the Blodgett Forest in the are not comparable to the fragmented forests Sierra in California. In their study, large mentioned above, although they might be com- diameter trees,used for singingand foraging,were parable to the corridors that Robbins (1979) rec- more important in summer, while dense canopy ommended to connect isolated woodland frag- cover for thermal protection was more important ments. in winter. Another difference between summer and win- Results of this project suggestthat the presence ter seasonswas seenin the comparisons between of some riparian vegetation is very important to large and small stands in open grasslands.The speciesrichness and total density, because con- greater richness and density in the small stands trol habitats had significantly lower diversity and in winter suggestthat either the large stands of fewer individuals than either large or small ri- riparian trees were less important to the birds in parian habitats. These findings are consistentwith the winter than in summer, or the open grass- those of Stevens et al. (1977) who found greater lands were more important than the riparian DISTRIBUTION IN RIPARIAN HABITATS 877

zonesin winter. The latter option seemsplausible zine sparrowspartitioned spacebased on the dis- becauseof high densities of several wintering em- tance to trees or shrubsthat could provide cover. berizid species,but it is not consistent with the In general, the type of adjacent upland vegetation low richness and density in the open, control had a greater impact on bird populations in win- plots. A possible explanation is that the grass- ter than in summer. The most consistent effect lands are important as a feeding area, but some was on speciesrichness of winter birds, with open riparian vegetation is important to provide cov- plots having a greater diversity of birds than er. The grasslandsmight have been less impor- wooded plots, in either large or small stands. tant in the summer because of the lower pro- Overall, the primary factor influencing bird portion of emberizids in the summer avifauna. speciesrichness and total density in summer was The type of adjacent upland vegetation has the dominant riparian tree species.In contrast, also been shown to have significant effects on adjacent upland vegetation had a more impor- birds in riparian vegetation. Carothers (1977) re- tant impact on the diversity and abundance of ported higher densities of breeding birds in cot- wintering bird populations. The sizesof riparian tonwood standsadjacent to agricultural lands than stands appeared to be less important than either in stands surrounded by pinyon-juniper, oak, or riparian tree species or the nature of adjacent chaparral vegetation. However, Conine et al. upland vegetation. (1978) found that certain riparian bird species were lost from a riparian community following PATTERNS OF BIRD SPECIES agricultural encroachment, even though overall DISTRIBUTION-SUMMER densities were higher. Szaro and Jakle (1985) Cluster analysis of habitats based on breeding found that bird densities decreasedsharply from bird densities (Fig. 6) revealed a clear distinction a riparian zone to a surrounding desert upland. between montane, heavily wooded habitats vs. They also found that the riparian bird commu- lowland and foothill habitats. Within the lower nity made a substantial contribution to bird pop- elevation habitats, there were breaks between ulations in the desertuplands, but the desertbird those dominated by smaller riparian trees (ash, community contributed very little to riparian desert willow, and walnut) and those dominated populations. by sycamoresand cottonwoods. This pattern in- In our study area, adjacent upland vegetation dicates that the bird species were distributing had significant relationships with richness and themselves partially according to an elevational density. In open grasslands,isolated riparian trees gradient and partially according to the riparian of almost any size will act as a focal point for vegetation. This result is consistent with the pat- bird nesting and foraging activities, leading to terns of speciesdistributions among habitats re- very high local densities around the central trees ported by Anderson et al. (1977) in the Colorado that were not representative of the surrounding River Valley, althoughthey were working at much grassland. For example, Cassin’s and Western lower elevations in areas with local habitat di- kingbirds were common nesting speciesin small versity. cottonwood and sycamoreplots. In winter, flocks The dendrogram based on breeding bird den- of fiingillids and emberizines foraged in grass- sities correlations with vegetation characteristics lands around the riparian trees. Because of the (Fig. 4) showsa definite pattern of speciesgroups small size of these stands (often just a single tree which is consistent with the pattern of habitats. or a very small grove), tinchesand sparrowscould Groups of high montane bird specieswere dis- be feeding in the grasslandand still be very close tinct from the groups of lowland or foothill spe- to the center tree of a study plot. The low winter cies. Some bird speciesgroups were easily as- density and richness of the open control plots signed to specific habitats, while other species compared with the small open plots suggeststhat were widely distributed in many habitats. It might the riparian vegetation was providing some es- be appropriate to consider some of these groups sential requirement for thesebirds. It seemsmost as distinctive avian communities that were co- likely that the trees provided cover in the form incident with specific vegetation habitats. of perching sitesbetween foraging bouts and pos- Within the Huachuca Mountains riparian hab- sibly additional protection from predators. This itats, there were a number of bird speciesthat finding is consistent with that of Pulliam and were restricted to or reached their greatestden- Mills (1977) who reported that several emberi- sities in a large cottonwood cienega.This habitat 878 THOMAS R. STRONG AND CARL E. BOCK

was dominated by very large cottonwood trees, Buff-breastedFlycatcher, Steller’s Jay, American with abundant water, surrounded by open grass- Robin, Grace’s Warbler, and Yellow-eyed Junco. land with some mesquite. Some typical bird spe- Recognition of a cottonwood cienega com- ciesin this community included the Yellow-billed munity, a grassland/desertscrub community, a Cuckoo, Vermilion Flycatcher, and Yellow War- montane canyon community, and a montane co- bler. Several species, such as the Gila Wood- niferous forest community emphasizes the im- pecker, Yellow-breasted Chat, and Abert’s Tow- portance of elevation in the distribution of breed- hee, appearedto be restricted to this habitat, and ing bird speciesin the Huachuca Mountains. This were rarely, if ever, recorded in other areas. distribution of bird speciestends to parallel the Gambel’s Quail was commonly recorded in this elevational distribution of dominant riparian habitat, but it did not appear to be dependent plant speciesdescribed by Paseand Layser (1977) on the cottonwoods. Rather, it seemed to prefer for arid and semi-arid environments of the the dry, mesquite habitat adjacent to the riparian Southwest, although many plants and birds in zone. The strong correspondence between bird the Huachuca Mountains showedfairly wide ele- speciesand cottonwood trees is consistent with vational tolerances, and many species over- the relationships between species richness and lapped in their distributions. density and cottonwoods. In a cottonwoodwil- Rice et al. (1984) found that tree speciescom- low habitat in the Colorado River Valley, Ro- position was very important in bird specieshab- senberg et al. (1982) found many insectivorous itat selection in riparian areas along the lower bird speciessharing similar resources,and in par- Colorado River. They reported that birds were ticular, several speciesconcentrated on cicadas, selecting for specific trees rather than showing a superabundant, seasonally predictable re- an avoidance of other trees. Wiens and Roten- source. Although we collected no data on for- berry (198 1) also reported many significant cor- aging behavior or food availability, it seemsrea- relations between bird species’ abundances and sonable that a stand of very large cottonwood the coverage of shrub speciesin a shrubsteppe trees along an active stream would provide food environment. They found both positive and neg- resourcesfor a large variety of speciesand quan- ative correlations, indicating that birds were tities adequate for high densities of birds. choosingsome shrub specieswhile avoiding oth- A grassland/desert scrub bird community ers. The relationships that we observed between (branch C, Fig. 4) included many speciescom- certain bird speciesand specificriparian tree spe- mon in the lowland, open grassland areas with cies,especially cottonwoods and sycamores,again some mesquite and desert willow, and relatively demonstrate the importance of measuring tree little water. Some representative bird speciesin- speciescomposition in habitat analysis studies, clude the Scaled Quail, Greater Roadrunner, as recommended by Rice et al. (1984). Western Kingbird, Chihuahuan Raven, Each described avian assemblage was based Wren, Canyon Towhee, and Botteri’s Sparrow. on a set of bird speciessharing similar patterns A montane canyon bird community (branch of distribution and similar correlations with veg- G, Fig. 4) was common in upper elevation can- etation. However, the bird species mentioned yons dominated by sycamoresand maples, with above were usually not the most common species fairly denseoak woodlandson upper slopes.These within any given habitat. There were several spe- canyons were fairly broad and contained reliable cies that occurred in relatively high densities in seasonal or permanent water. This community a wide variety of habitats. These species were included several of the specieswhose range with- often numerically dominant, but they could not in the is restricted to mountains be easily assignedto any single community, and in this comer of Arizona, suchas the Magnificent they were not indicators of specificassemblages. Hummingbird, Elegant Trogon, Strickland’s Some examples of thesewidespread specieswere Woodpecker, Dusky-capped Flycatcher, and the Western Wood-, Ash-throated Fly- Painted Redstart. catcher, Cassin’s Kingbird, Gray-breasted Jay, A montane, coniferous forest bird community Bewick’s Wren, Lucy’s Warbler, Brown-headed (branch H, Fig. 4) was found in areas that in- Cowbird, and House Finch. This observed pat- cluded the highest of our study plots, dominated tern is consistent with the results of Wiens and by a relatively open pine forest. Some typical Rotenberry (198 1) who found that the most species of this group were the Greater Pewee, widely distributed bird speciesshowed little cor- DISTRIBUTION IN RIPARIAN HABITATS 879 relation with habitat features, but specieswith the winter, but they were primarily found in the localized distributions tended to have much mountains during the summer. strongerassociations with habitat characteristics. Holmes et al. (1986) found that different bird ACKNOWLEDGMENTS speciesshowed different patterns of abundance We would like to acknowledgethe followingpeople over a 15-year period in the Hubbard Brook Ex- whoprovided support for this project:Robert C. Szaro perimental Forest, suggestingthat each species of the U.S. ForestService; Clark Derdeynand Randy was responding to different factors in the envi- Brelandof theFort HuachucaWildlife 05ce; and Mark ronment. Some speciesseemed more closely tied Stromberg of the Appleton-Whittell Research Ranch. This project was supported in part by Cooperative to local habitat variables, while others were af- Aareement No. 28-C3-279 with the Rockv Mountain fected more by regional or global scale events. F&est and Range Experimental Station of:the United They took a very broad view of avian community States Forest Service, by the National Audubon So- structure, with many different factors interacting ciety, and by University of Colorado ResearchFellow- to determine the assemblageof birds at any par- ships. ticular time. Although our study was over a much shorter time period, our results support this view LITERATURE CITED of avian communities. ANDERSON, B. W., A. E. HIGGINS,AND R. D. &MART. 1977. Avian use of salt cedar communities in the PATTERNS OF BIRD SPECIES Lower Colorado River Valley, p. 128-136. In R. DISTRIBUTION-WINTER R. Johnson and D. A. Jones [tech. coords.], Im- portance, preservation,and managementof ripar- During the winter season, different patterns of ian habitats: a symposium. U.S.D.A. For. Serv. bird species and vegetation relationships were Gen. Tech. Rep..RM-43. Fort Collins, CO. apparent. The dendrogram of habitats based on ANDERSON. B. W.. AND R. D. OHMART. 1977. Vea- etation structure and bird use in the Lower Cz- winter bird densities (Fig. 7) showed a very clear orado River Valley, p. 23-34. In R. R. Johnson break between wooded and open habitats. This and D. A. Jones [tech. coords.], Importance, pres- was somewhat similar to the summer habitat ervation, and management of riparian habitats: a pattern, but the separation was more clearly re- symposium. U.S.D.A. For. Serv. Gen. Tech. Rep. lated to upland vegetation than to elevation. This RM-43. Fort Collins, CO. BLAKE,J. G. 1986. Species-arearelationship of mi- contrast suggeststhat dominant riparian tree spe- grantsin isolated woodlots in east-centralIllinois. cies are lessimportant than upland vegetation in Wilson Bull. 98:291-296. determining winter bird communities, as op- BLAKE,J. G., AND J. R. KARR. 1984. Speciescom- posedto the importance of riparian trees in sum- position ofbird communities and the conservation of large versus small forests. Biol. Conserv. 30: mer bird communities. The dendrogram of win- 173-187. ter bird speciesbased on vegetation correlations BOCK,C. E., AND J. H. Bocx. 1984. Importance of (Fig. 5) gave patterns that were consistent with sycamoresto riparian birds in southeasternAri- the pattern of habitats. Groups of specieswere zona. J. Field Omithol. 55:97-103. clearly associatedwith wooded areas or open ar- CARoTHERs,S. W. 1977. Importance, preservation, and managementof riparian habitat: an overview, eas. However, within each main division, there p. 2-4. In R. R. Johnson and D. A. Jones [tech. were no obvious associations with specific ri- coords.], Importance, preservation, and manage- parian habitats. ment of ri~arian habitat: a svmoosium. U.S.D.A. Some permanent resident species,such as the Forest Service Gen. Tech. Rep. RM-43. Fort Col- lins, CO. Abert’s Towhee, were found in the same habitat CAROTHERS,S. W., R. R. JOHNSON, AND S. W. in both summer and winter, but others, like the Amcxmo~. 1974. Population and social orga- Gila Woodpecker, were found in a greatervariety nization of southwesternriparian birds. Am. Zool. of habitats in the winter. This observation for 14:97-108. Abert’s Towhee is in contrast to Anderson and CANINE,K. H., B. W. ANDaRsON,R. D. OHMAaT,AND J. F. DRAKE. 1978. Responsesof riparian species Ohmart (1977) who found this speciesto be more to agriculturalconversions, p. 248-262. In R. R. of a habitat specialistduring winter than in sum- Johnson and J. F. McCormick [tech. coords.], mer. Other species that were year-round resi- Strategies for protection and management of dents within the study area were found in dif- floodplain wetlands and other riparian ecosys- tems. U.S.D.A. For. Serv. Gen. Tech. Rep. WO- ferent elevation zones in the winter. For example, 12. Washington, DC. the House Wren and the Rufous-sided Towhee How, R. T., T. W. SHERRY,AND F. W. STURGES. were found in a variety of lowland habitats in 1986. Bird community dynamics in a temperate 880 THOMAS R. STRONG AND CARL E. BOCK

deciduous forest: long-term trends at Hubbard Graaf and K. E. Evans lcomps.],_ _. Management of Brook. Ecol. Monogr.56:201-220. north central and northeasternforests fornongame Hurro, R. L. 1985. Seasonalchanaes in the habitat birds. U.S.D.A. For. Serv. Gen. Tech. Reo. NC- distribution of transient insec&orous birds in 5 1. St. Paul, MN. southeasternArizona: competition mediated?Auk ROHLF,F. J., J. KISHPAUGH, AND D. Kmtc. 1972. Nu- 102:12&132. merical system of multivariate statis- JAMES,F. C., ANDH. H. SHUGART,JR. 1970. A quan- tical programs(NT-SYS). StateUniv. ofNew York, titative method of habitat description. Audubon Stony Brook. Field Notes 24~727-736. ROSENBERO,K. V., R. D. OHMART, AND B. W. JOHNSON, R. R., L. T. HAIGHT, AND J. M. SJMPWN. AND-N. 1982. Community organization of 1977. Endangeredspecies vs. endangeredhabitat: riparian breeding birds: responseto an annual re- a conceut, D. 68-79. In R. R. Johnson and D. A. sourcepeak. Auk 99:260-274. Jones [tech. coords.], Importance, preservation, STAMP, N. 6. 1978. Breeding birds of riparian wood- and management of riparian habitats: a sympo- lands in south-centralArizona. Condor 80:64-7 1. sium. U.S.D.A. For. Serv. Gen. Tech. Rep. RM- STEVENS,L. E., B. R. BROWN,J. M. SIMPSON,AND R. 43. Fort Collins, CO. R. JOHNSON.1977. The importance of riparian KARR, J. E., AND K. E. F~~EMARK. 1983. Habitat habitat to migrating birds, p. 156-164. In R. R. selectionand environmental gradients:Dynamics Johnson and D. A. Jones [tech. coords.], Impor- in the “stable” tropics. Ecology 64: 1481-1494. tance, preservation, and management of riparian KNOPF,F. L. 1985. Significanceof riparian vegetation habitats: a symposium. U.S.D.A. For. Serv. Gen. to breeding birds across an altitudinal cline. Tech. Rep. RM-43. Fort Collins, CO. U.S.D.A. For. Serv. Gen. Tech. Rep. RM-120: STRONG,T. R. 1987. Bird communities in the ripar- 105-l 11. Fort Collins, CO. ian habitats of the Huachuca Mountains and vi- MORRISON,M. L., K. A. Wrrn, AND I. C. Tr~oss~. cinity in southeasternArizona. Ph.D.diss.. Univ. 1986. The structure of a forest bird community of Colorado, Boulder. during summer and winter. Wilson Bull. 98:214- SZARO,R. C. 1980. Factors influencing bird popu- 230. lation is southwesternriparian forests,p. 403-4 18. OPDAM,P., G. RUSDUK,AND F. HU~TNGS. 1985. Bird In R. M. DeGraff [tech. coord.], Management of communities in small woods in an agricultural western forestsandgrasslands for nongame birds. landscane:effects of area and isolation. Biol. Con- U.S.D.A. For. Serv. Gen. Tech. Ren. INT-86. Oe- serv. 34:333-352. den, UT. PASE, C. P., AND E. F. L~vsaa. 1977. Classification SZARO,R. C., AND M. D. JAKLE. 1982. Comparison of riparian habitat in the southwest,p. 5-9. In R. of variable circular-plot and spot-map methods in R. Johnson and D. A. Jones [tech. coords.], Im- desert riparian and scrub habitats. Wilson Bull. portance, preservation,and managementofripar- 94:546-550. ian habitats: a svmwsium. U.S.D.A. For. Serv. Smo, R. C., AND M. D. JAKLE. 1985. Avian use of Gen. Tech. Rep.-RM-43. Fort Collins, CO. a desert riparian island and its adjacentscrub hab- Parrr, D. R., K. E. Pa-rrr, ANDT. C. GRUBB, JR. 1985. itat. Condor 87:511-519. On atmospheric moisture as a factor influencing TAYLOR,C. 1980. The Coppery-tailed Trogon: Ari- distribution of breeding birds in temperate decid- zona’s “bird of paradise.”Borderland Publications, uous forest. Wilson Bull. 97:88-96. Portal, AZ. PULLIAM,H. R., AND G. S. Mars. 1977. The use of VOWER, J., AND L. V. Rrrrua. 1985. A comparison space by wintering sparrows. Ecology 58:1393- of transectsand point counts in oak-pine wood- 1399. lands of California. Condor 87147-68. REYNOLDS,R. T., J. M. Scorr, AND R. A. Nussn~u~. WHITCOMB,R. F. 1977. Island biogeography and 1980. A variable circular-plot method for esti- “habitat islands” of easternforest. I. Introduction. mating bird numbers. Condor 82:309-3 13. Am. Birds 313-5. Rrca, J., B. W. ANDERSON,AND R. D. GHMAaT. 1984. Wws, J. A., ANDJ. T. ROTFZNBERRY.1981. Habitat Comparison of the importance of different habitat associationand community structure of birds in attributes to avian community organization. J. shrubsteppeenvironments. Ecol. Monogr. 5 1:2I- Wildl. Manage. 48:895-911. 41. ROBBINS,C. S. 1979. Effects of forest fragmentation on bird populations, p. 198-212. In R. M. De- DISTRIBUTION IN RIPARIAN HABITATS 881

APPENDIX A. Habitat list. List of 25 habitat types used in the multivariate statisticalanalyses. Numbers correspondwith those used in Appendices B and C to indicate which habitats were utilized by each species.

Upland NUmk Tree spcies Standsize vegetation

1 Control Wooded 2 Control Open 3 Ash Large wooded 4 Ash Large Open Ash Small wooded : Ash Small Open Cottonwood Large Wooded : Cottonwood Large Open 9 Small Wooded 10 Cottonwood Small Open 11 Desert willow Large Wooded 12 Desert willow Large Open 13 Desert willow Small wooded 14 Desert willow Small Open 15 Maple Large Wooded 16 Maple Small Wooded 17 Sycamore Large Wooded 18 Sycamore Large Open Sycamore Small Wooded :: Sycamore Small Open 21 Walnut Large Wooded 22 walnut Large Open 23 Walnut Small Wooded 24 walnut Small Open 25 Willow Small Open 882 THOMAS R. STRONG AND CARL E. BOCK

APPENDIX B. Specieslist-summer. List of 87 speciesused in the multivariate statistical analyses.Species numbers correspond to those used in Figure 4. Habitat numbers for each speciescorrespond to the habitats listed in Appendix A and indicate those in which the species’ density was greater than one individual per 10 ha. An asterisk (*) indicates the habitat of maximum density for a specieswhose density was less than one individual per 10 ha in all habitats.

speciesnumber Habitats

1 Mallard Anas platyrhynchos 8, 25 2 Turkey Vulture Cathartesaura 8+ 3 Red-tailed Hawk Buteojamaicensis 9* 4 American Kestrel Falco sparverius 10,20 5 Montezuma Quail Cyrtonyx montezumae 5, 10, 11, 16, 17, 20 6 Scaled Quail _ Callipepla squamata 6, 22, 24 7 Gambel’s Ouail Callipepla gambehi 8, 24 8 Killdeer _ Charadrius voctferus 6, 10, 11, 14, 20, 24 9 Band-tailed Pigeon ColumbafasCata * 10 White-winged Dove Zenaida asiatica 155 6 7 8 9 10 11 14, 17, ’ ’ 18, io,‘2i, i2,i3, is 11 Mourning Dove Zenaida macroura 2, 4, 6, 7, 8, 9, 10, 11, 12, 14, 18, 20, 21, 22, 23, 24 12 Common Ground-Dove Columbina passerina 8* 13 Yellow-billed Cuckoo Coccyzusamericanus 21* 14 Greater Roadrunner Geococcyxcalifornianus 12; 15 Great Homed Owl Bubo virginianus 18,23 16 Common Nighthawk Chordeilesminor 7* 17 White-throated Swift Aeronautessaxatalis 15, 16 18 Magnificent Hummingbird I Eugenesfulgens 15, 17, 19, 21 19 Black-chinned Hummingbird Archilochusalexandri 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25 20 Broad-tailed Hummingbird Selasphorusplatycercus 1, 15, 16, 17, 19, 21 21 Elegant Trogon Trogon elegans 22 Acorn Woodpecker Melanerpesformicivorus i77 9 10 11 13, 17, 19,20, ‘2i, i2,i3, is 23 Gila Woodpecker Melanerpesuropygiahs Ladder-backed Woodpecker Picoidesscalaris !. 10, 22. 24 ;: Strickland’s Woodpecker Picoidesstricklandi 1617 26 Northern Flicker Colaptesauratus 3, 5, 7, 8, 10, 11, 17, 19, 20, 21, 22, 23 27 Greater Pewee Contopuspertinax 19 28 Western Wood-Pewee Contopussordid&s 1, 3, 7, 8, 9, 10, 11, 15, 17, 18, 19, 21, 24 29 Buff-breastedFlycatcher Empidonax fulvifiions 1, 19 30 Black Phoebe Sayornis nigricans 5, 6, 7, 8, 9, 20, 25 31 Say’s Phoebe Sayornis saya 6, 7, 11 Vermilion Flycatcher Pyrocephalusrubinus 6, 8, 10, 14, 18, 20, 25 :: Dusky-capped Flycatcher Myiarchus tuberculife 1, 3, 7, 16, 17, 18, 19 34 Ash-throated Flycatcher Myiarchus cinerascens 1, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 17, 18, 19, 20, 21, 22, 23,24 Brown-crested Flycatcher Myiarchus tyrannulus 8, 10 :; Sulphur-bellied Flycatcher MyiodynastesIuteiventris 3, 15, 17 37 C&sin’s Kingbird Tyrannus voctferans 1, 3, 4, 6, 7, 8, 9, 10, 11, 13, 17, 18, 19, 20, 21, 22, 23, 24, 25 38 Western Kingbird Tyrannus verticalis 6,9, 10, 14, 18, 19, 20, 22, 24, 25 39 Violet-green Swallow Tachycinetathalassina 40 Barn Swallow Hirundo rustica is9 10 24 41 Steller’s Jay stelleri lb,21 ’ 42 Gray-breasted Jay Aphelocomaultramarina 1, 3, 5, 9, 11, 13, 15, 16, 17, 19.23 43 Chihuahuan Raven Corvuscryptoleucus 18* DISTRIBUTION IN RIPARIAN HABITATS 883

APPENDIX B. Continued.

speciesnumber Habitats

Bridled Titmouse Parus wollweberi 1, 3, 5, 7, 9, 11, 13, 15, 16, 17, 18, 19, 20, 21, 23 Bushtit Psaltriparusminimus 1, 3, 5, 7, 8, 9, 11, 13, 15, 17, 19, 20, 21, 23 46 White-breasted Nuthatch Sitta carolinensis 1, 3, 7, 17, 23 47 Cactus Wren Campylorhynchusbrunneicapi1Iu.s 4, 12 48 Rock Wren Salpinctesobsoletus 2. 11 49 Canyon Wren Catherpesmexicanus 16 50 Bewick’s Wren Thryomanesbewickii 1, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 16, 17, 18, 19, 20, 21, 22, 23, 24 51 House Wren Troglodytesaedon 17 52 Eastern Bluebird Sialia sialis 9, 11, 17, 21, 23 53 Hermit Thrush Catharus guttatus 172 54 American Robin Turdus migratorius 1, 7, 9, 15, 17, 19,21 55 Northern Mockingbird Mimus polyglottos 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 18, 20, 21, 22, 23, 24 56 Curve-billed Thrasher Toxostomacurvirostre 8 51 Phainopepla Phainopepla nitens 4, 5, 7, 8, 10, 13 58 LoggerheadShrike Lanius Iudovicianus 10 59 Solitary Vireo Vireo solitarius 1, 3, 15, 17, 19, 21 60 Virginia’s Warbler Vermivora virginiae 16,23 61 Lucy’s Warbler Vermivora luciae 3, 4, 7, 8, 9, 10, 11, 12, 14, 18, 21, 22, 23, 24 Yellow Warbler Dendroicapetechia 7, 8, 10 Black-throated Warbler Dendroica nigrescens 1, 3, 10, 16, 17, 19, 20 Grace’s Warbler Dendroicagraciae 1, 19 Common Yellow-throat Geothlypistrichas 2, 5, 6, 7, 9, 10, 14, 20, 21, 22, 24,25 66 Red-faced Warbler Cardellina rubrtfions 16 Painted Redstart Myioboruspictus 1, 3, 15, 16, 17, 19, 21 :: Yellow-breasted Chat Icteria virens 8 69 Hepatic Tanager Piranga flava 15,17 70 Summer Tanager Piranga rubra 3, 4, 5, 7, 8, 9, 11, 18 71 Black-headedGrosbeak Pheucticusmelanocephalus 1, 3, 5, 7, 9, 15, 16, 17, 19, 21, 23 72 Blue Grosbeak Guiraca caerulea 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 17, 18, 19, 20, 21, 22, 23,25 73 Rufous-sided Towhee Pipilo erythrophthalmus 7, 15, 16, 19, 21 74 Canyon Towhee Pipilo fuscus 4, 6, 13, 14, 18, 20, 21, 22, 24 15 Abert’s Towhee Pipilo aberti 8 76 Botteri’s Sparrow Aimophila botterii 2,4, 6, 10, 14, 22, 24 77 Rufous-crowned Sparrow Aimophila rujiceps 1,2,3,4, 5, 10, 11, 12, 13, 17, 18, 20, 21, 22, 23, 24 18 Lark Sparrow Chondestesgrammacus 5, 9, 10, 22, 23 79 Black-throated Sparrow Amphispiza bilineata 2 80 Yellow-eyed Junco Juncophaeonotus 15,19 81 Red-winged Blackbird Agelaiusphoeniceus 7, 9, 10, 14, 25 82 Eastern Meadowlark Sturnella magna 2, 4, 6, 10, 12, 14, 22, 23, 24, 25 83 Brown-headed Cowbird Molothrus ater 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 16, 17, 18, 19, 20, 21, 22, 24, 25 Northern Oriole Icterus galbula 6. 8, 10, 18, 20, 24 Scott’s Oriole Icterusparisorum 5; 20,24 House Finch Carpodacusmexicanus 3, 4, 5, 6, 7, 8, 10, 11, 12, 13, 14, 17, 18, 20, 21, 22, 23, 24, 25 87 Lesser Goldfinch Card&is psaltria 4, 7, 8, 11, 17, 18, 19 884 THOMAS R. STRONG AND CARL E. BOCK

APPENDIX C. Specieslist-winter. List of 60 bird speciesused in the multivariate statisticalanalyses. Numbers correspond with those used in Figure 5. Habitat numbers for each speciescorrespond to the habitats listed in Appendix A and indicate those in which the species’ density was greater than one individual per 10 ha.

1 Northern Harrier Circuscyaneus 12 Red-tailed Hawk Buteojamaicensis 10, 20, 24 ; American Kestrel Falco sparverius 7, 8, 20 4 Montezuma Quail Cyrtonyx montezumae 2, 12, 16 5 Scaled Quail Callipeplasquamata 2, 20 6 Gambel’s Quail Callipeplagambelii Killdcer Charadriusvoctferus t 10 14 25 : Common Snipe Gallinagogallinago 6: 10: 14: 20, 25 9 Mourning Dove Zenaida macroura 6, 10, 14, 20, 25 10 Greater Roadrunner Geococcyxcaltfornianus 9, 20, 24 11 Acorn Woodpecker Melanerpesformicivorus 3, 7, 9, 13, 17, 19, 20 12 Gila Woodpecker Melanerpesuropygialis 6, 8, 21 13 Yellow-bellied Sapsucker Sphyrapicusvarius 1, 3, 5, 7, 8, 15, 16, 17, 21 14 Williamson’s Sapsucker Sphyrapicusthyroideus 1, 19 15 Ladder-backed Woodpecker Picoidesscalaris 3, 4, 7, 8, 10, 18, 22, 24, 25 16 Strickland’s Woodpecker Picoidesstricklandi 13.21 17 Northern Flicker Colaptesauratus 7,8, 10, 11, 13, 14, 16, 17, 19, 20, 22, 24 18 Black Phoebe Sayornisnigricans 5, 6, 7, 8, 9, 10, 11, 14, 18, 20, 25 19 Say’s Phoebe Sayornissaya 3, 4, 6, 7, 8, 14, 20, 25 20 Homed Lark Eremophila alpestris 2 Steller’s Jay Cyanocittastelleri 9 4: Gray-breasted Jay Aphelocomaultramarina 1, 7, 11, 13, 15, 16, 17, 20, 21 23 Chihuahuan Raven CorvuscIyptoleucus 10, 20, 22, 23 24 Bridled Titmouse Parus wollweberi 1, 3, 5, 7, 9, 11, 13, 15, 16, 17, 19,23 25 Verdin Auriparusflaviceps 2, 4, 12, 18, 22, 23, 24, 25 26 Bushtit Psaltriparusminimus 1, 3, 5, 6, 7, 9, 11, 13, 14, 15, 16, 17, 20, 21, 22, 23 27 White-breasted Nuthatch Sitta carolinensis 3,7,9, 11, 13, 15, 17,21,22 28 Cactus Wren Campylorhynchusbrunneicapillus 18 29 Rock Wren Salpinctesobsoletus 2, 5, 8, 11, 12, 17, 22, 23 30 Canyon Wren Catherpesmexicanus 15, 16, 21 31 Bewick’s Wren Thryomanesbewickii 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 17, 18, 19, 20, 21, 22, 23, 24.25 House Wren Troglodytesaedon 9,24 Ruby-crowned Kinglet Reguluscalendula 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 25 Eastern Bluebird Sialia sialis 7 Western Bluebird Sialia mexicana 1, 5, 17, 21 Hermit Thrush Catharusguttatus 1, 16, 19 American Robin Turdus migratorius 7, 18 Curve-billed Thrasher Toxostomacurvirostre 3, 4, 8, 12 LoggerheadShrike Lanius ludovicianus 4, 6, 7, 9, 10, 18 Yellow-rumped Warbler Dendroicacoronata 1, 2, 3, 4, 6, 7, 8, 9, 10, 11, 12, 14, 17, 18, 20, 21, 22, 24, 25 41 Pyrrhuloxia Cardinalis sinuatus 4, 8, 13, 18, 21 42 Green-tailed Towhee Pipilo chlorurus 8 43 Rufous-sided Towhee Pipilo erythrophthalmus 1, 3, 5, 7, 8, 9, 11, 13, 16, 18, 19,20 44 Canyon Towhee Pipilo furcus 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 17, 18, 20, 21, 22, 23,

45 Abert’s Towhee Pipilo aberti 8,294 DISTRIBUTION IN RIPARIAN HABITATS 885

APPENDIX C. Continued.

speciesnumber Habitats

46 Rufous-crowned Sparrow Aimophila rujiceps 1, 2, 3, 4, 7, 9, 10, 13, 14, 18, 19, 20, 21, 22, 23, 24 47 Chipping Sparrow Spizella passerina 1, 3, 5, 6, 7, 9, 11, 12, 13, 14, 17, 18, 20, 21, 22, 23, 24 48 Vesper Sparrow Pooecetesgramineus 2, 4, 6, 8, 9, 12, 14, 18, 20, 24, 25 49 song sparrow Melospiza melodia 3, 4, 6, 8, 10, 14, 18, 20, 24 50 Lincoln’s Sparrow Melospiza lincolnii 3, 9, 10, 11, 14, 18, 19, 20, 21, 24 51 White-crowned sparrow Zonotrichia leucophrys 2, 4, 6, 8, 9, 12, 14, 18, 20, 21, 22, 23, 24 52 Dark-eyed Junco Junco hyemalis 1, 3, 4, 5, 7, 9, 11, 13, 15, 16, 17, 18, 19, 20, 21, 23 53 Yellow-eyed Junco Juncophaeonotus 1, 3, 9, 15, 17, 19,21 54 Red-winged Blackbird Agelaiusphoeniceus 6, 8, 9, 10, 24, 25 Eastern Meadowlark Sturnella magna 2, 4, 10, 12, 22, 24, 25 :z C&sin’s Finch Carpodacuscassinii 3, 11, 17 57 House Finch Carpodacw mexicanus 3, 4, 6, 8, 10, 12, 17, 18, 20, 21 58 Pine Siskin Card&is pinus 5, 7, 8, 9, 10, 11, 14, 17, 20, 21 59 Lesser Goldfinch Card&is psaltria 8, 9, 12, 18, 19 60 American Goldfinch Card&is tristis 4, 6, 8, 10, 11, 17, 18, 20

APPENDIX D. List of physical and vegetation characteristicsmeasured on 132 riparian study plots and used in statisticalanalyses.

Stand size--Ordinal Maple density, trees/ha Distance to next stand-Ordinal Maple basal area, m2/ha Presenceof water-Ordinal Sycamoredensity, trees/ha Upland tree type--Ordinal Sycamorebasal area, m2/ha Upland tree density- Ordinal Walnut density, trees/ha Elevation, meters Walnut basal area, m2/ha Horizon angle-east Willow density, trees/ha Horizon angle-west Willow basal area, m*/ha Canopy height, meters Juniper density, trees/ha Riparian canopy, % Juniper basal area, m2/ha Upland canopy, % Madrone density, trees/ha Garss-herb cover, % Madrone basal area, m*/ha Shrub canopy, total, % Mesquite density, trees/ha Sapling canopy, total, % Mesquite basal area, ml/ha Upland tree average distance, meters Oak density, trees/ha Tree density, total, trees/ha Oak basal area, m*/ha Tree basal area, total, m2/ha Pine density, trees/ha Number of trees > 30 cm and < 60 cm dbh Pine basal area, m2/ha Number of trees ~60 cm dbh Pinyon density, trees/ha Ash density, trees/ha Pinyon basal area, m*/ha Ash basal area, m2/ha Snag density, trees/ha Cottonwood density, trees/ha Snag basal area, m2/ha Cottonwood basal area, m2/ha Desert willow density, trees/ha Desert willow basal area, m*/ha