BOTANICAL CHARACTERISTICS OF AUTUMNAL BLACKBIRD ROOSTS IN CENTRAL OHIO 1

Transcription

1 Copyright Ohio Acad. Sci. OOO-OO//OOO-OI $.00/0 BOTANICAL CHARACTERISTICS OF AUTUMNAL BLACKBIRD ROOSTS IN CENTRAL OHIO MICK MICACCHION and T. W. TOWNSEND, Division of Fisheries and Wildlife Management, School of Natural Resources, The Ohio State University, Columbus, OH 0 and the Ohio Agricultural Research and Development Center, Wooster, OH 6 ABSTRACT. Vegetative parameters of autumn blackbird roosts in central Ohio were measured. Common characteristics of roosts included young trees (x =.0 years, SD =.6), high densities (x = 776 trees/ha, SD = ), and thick canopies. Plant communities were adapted to moist sites and included American elm {Ulmus amencana), silver maple ( saccharinum), green ash (Fraxinus pennsy/vanica), red maple { rubrum), sugar maple ( saccbarum) and black cherry (Prunus serotina). Tree density, twig height, and twig index were significantly related to numbers of birds per ha. OHIOJ. SCI. (): -, INTRODUCTION Red-winged blackbirds (Agelaius phoeniceus), common grackles (Quiscalus quiscala), brown-headed cowbirds (Molothrus ater), and starlings (Sturnus vulgaris) spend nights in concentrations at common roost sites in late summer and fall. s often contain thousands of birds, and noise and droppings are often objectionable to people living nearby. Blackbirds can cause economic losses (Dolbeer et al. 7), pest problems (Meanley 7), hazards to aircraft (Bloekpoel 76), and may be dangerous to human health (Tosh et al. 70). s constitute points where management to relieve blackbird problems could be concentrated. Little is known about the vegetation comprising tree roost sites. Several previous studies found roosts situated in single species stands such as mature, open grown live oak (Quercus virginiana) (Good and Johnson 7), a plantation of 0-year-old loblolly pine (Pinus taeda) (Francis 76), and a plantation of white pine (Pinus strobus) (Kelty and Lustick 77). Mature 'Manuscript received July and in revised form 0 September (#-). Send reprint requests to Dr. T. W. Townsend, 7 Kottman Hall, 0 Coffey Rd., The Ohio State Univ., Columbus, OH 0. Norway maple ( platanoides) and sycamore (Platanus occidentals) were used for urban roosting in Pennsylvania (Jumber 6). In an urban Iowa roost, large maples ( spp.) were used by blackbirds (Bliese ). New Jersey roosts appear to be typified by young, early successional hardwood stands of diverse composition (Lyon 7, Lyon and Caccamise ). However, most of these studies have not involved detailed measurements of vegetation comprising roosts, and no study has considered autumn roosts in Ohio. Detailed vegetation information is necessary if roosting behavior is to be fully understood and modified when necessary. The present study was undertaken to define quantitatively and qualitatively the vegetation characteristics of fall roosts utilized by blackbirds in central Ohio. STUDY SITE Efforts to locate blackbird roosts were concentrated in an area centered on the middle of Columbus, Ohio and extended 70 km in all directions; the circular area of study was approximately,000 km. The original vegetation type of this area was deciduous forest (Good 7). Presently, forested tracts are scattered among agricultural, residential, and industrial land holdings.

2 M. MICACCHION AND T. W. TOWNSEND Vol. METHODS AND MATERIALS Roads in the study area were driven in the evenings of August, September, and October 77 and 7, and flocks or flightlines were followed when possible. Fourteen roosts were located. populations were estimated weekly during September, October, and November until roost desertion; birds returning to the roost in evening were counted in blocks of 00. boundaries were determined by observing concentrations of bird droppings and mapped using the compass-traverse method (Mosby 7). A compensating polar planimeter was used to estimate areas of roosts. Tree species, density, frequency, and mean basal area were estimated for each roost using the pointcentered quarter method (Cottam and Curtis 6). Starting points for transects were randomly located on the roost perimeter. A point was sampled every 0 m along a randomly drawn azimuth. All roosts required at least transects, and number of points/roost ranged from to 0. Foliage height profiles (MacArthur and Horn 6) were measured in roosts before leaves had fallen. One roost was not located until after many leaves had fallen and therefore was not sampled. Sample points were chosen in the same manner as those for the point-centered quarter method. For each roost, 6 readings from a tripod-mounted -mm camera with -mm telephoto lens were taken at 0 separate sample points. The viewing screen of the camera had a -by- grid, and each intersection with canopy vegetation was recorded from the focusing scale of the lens for each of the 6 grid intersections. Twig height is the average distance from the camera lens at all sample points to the nearest piece of vegetation in the canopy. Twig index measures average number of pieces of vegetation touched by a straight vertical line passing through an average m of the canopy and indicates density. At each of roosts, 0 trees were sampled to determine age. A starting point on the roost perimeter, a compass direction and a number paces into the roost were randomly selected. The nearest tree to this point became the first sample and subsequent sample trees were randomly determined by a new compass direction and number of paces. An increment borer was used to extract a core at. m from ground, and annual rings were counted. RESULTS Maximum blackbird populations of the roosts ranged from 0,000 to 00,000 birds (table ). An approximation of species composition in the roosts was 0% grackles, 0% starlings, % red-winged blackbirds and % brownheaded cowbirds. A few American robins {Turdus migratorius) were observed at several roosts. Maximum roost populations were correlated with roost area (R = 0.). areas ranged from 0.0 to.7 ha (table ). Most roosts occupied the entire area or a large portion of an isolated woodlot. No roost was located in a stand larger than twice the size of the roost. All roosts were dominated by deciduous trees usually found on moist soils. American elm (Ulmus americana), silver maple { saccharinum), green ash (Fraxinuspennsylvanka), red maple { rubrum), sugar maple { saccharum), and black cherry {Prunus serotina) accounted for % of occurances (table ). Most roosts were comprised almost entirely of one,, or species. In all roosts, the most common tree species within each roost accounted for at least 0% of the stand. Trees dominated roost sites and only a sparse population of plants with little diversity existed at ground level. High tree densities were typical of roosts. The roosts ranged from 7 to,0 trees/ha with a grand mean of 776 trees/ha (SD = ) (table ). Mean basal area of roost trees ranged from.6 to.7 cm (table ). The roosts had a grand mean basal area of.7 cm (SD = 0.7). s with the largest mean basal areas showed the greatest variability in the range of basal areas. Larger basal areas were correlated with lower tree densities (R = 0.). Mean twig height ranged from - to. m (table ), and grand mean height was 6. m (SD =.6). Twig index ranged from. to.6 with a mean of. (SD = 0.) (table ). Variability within individual roosts was small, and most canopy vegetation occurred within a narrow vertical range. s with the highest canopies also had the most vertical depth in canopies. trees had a grand mean age of.0 years (SD =.6) (table ). Individual roosts ranged in mean age from.0 to 6. years (CV = -.); 67% of the roosts were between.0 and. years of age. Ages specified were minimum since corings were taken at. m from ground.

3 Ohio J. Sci. BLACKBIRD ROOST VEGETATION TABLE Blackbird populations, areas of roosts, and trees occurring most frequently in autumnal blackbird roosts in central Ohio, 777. Number Area (ha) Maximum Number Bird Trees Population Sampled Ulmus americana saccharinum Fraxinus pennsylvanica rubrum saccharum Prunus serotina Others ,000 60,000 00,000 0,000 00,000 7,000 00,000,000 0,000 0,000,000,000 0,000 0,000 TOTAL ,76 7 (7%) (6%) 0 ( %) 6 (%) 6 ( 6%) (7%) (%) 7 (7%) 0 (0%) ( %) (%) 7 (7%) (0%) 77 (0%) (6%) 70 (%) (%) (7%) (0%) 0 (%) (70%) (%) (%) (%) (%) ( %) ( %) (%) ( %) 6 (00%) 7 (%) (%) 0 (%) ( %) 0 ( %) ( %) ( 7%) ( %) 66 (%) 6 (6%) ( %) 7 ( %) ( %) ( %) 6 ( %) ( %) (%) 6 (%) ( %) 0 (%) ( %) ( %) (%) ( %) (%) (6%) ( %) (%) 7 (%) Some trees could have been years older than mentioned, but it is possible for all species encountered to reach. m in one year on the excellent soils of roosts. Mean basal areas of roosts were correlated with mean age (R = 0.). A multiple regression analysis was conducted using maximum birds per ha as the dependent variable and basal area, age, density, twig height, and twig index as the independent variables for each roost. Tree density (P < 0.0), twig height (P < 0.0), and twig index (P < 0.0) were significant in explaining the variation in birds per ha. When tree density, twig height, and twig index are considered together, they account for 7% of the variation in birds per ha. DISCUSSION The roosts share several distinguishable characteristics: the dominant species were deciduous trees usually found on moist soils; stand age was young and homogeneous; tree density was high, and tree canopies occurred within a narrow vertical range and were dense. Tree density, twig height, and twig index appear to be important factors influencing the number of birds attracted to roosts. These factors define a dense canopy which may be important in creating an advantageous microclimate (Francis 76, Kelty and Lustick 77) and/or may allow congregations of birds favorable for exchange of information about foraging loca-

4 M. MICACCHION AND T. W. TOWNSEND Vol. TABLE Vegetation parameters for fall blackbird roosts in central Ohio, 777. Number x Age (years) n = 0 x Density (trees/ha) X Basal area (cm ) n for Density and Basal Area x Twig Height (m) Twig Index i(.66)*. i( 6.).0 i(.). ('.).0 < (.).0 i(.) 0.0 i(.0). < (.) 6. '(.0) 6. '(.).70 i{.6) 0. { i.),0,, 0,,00 7,,7,7,,7,,07 6,7 6,. 6 (.06). ( 7.0). 7 (.). OK.0). 0 (.0) 6. (.). 00 ( 0.6). 7 ( 60.). 0 (.0). (.). 7 (.). (.7). 6 (.). 07 (.) (. ( 6.0 ( 6.6 (.0 (.7 (. ( 7.0 (. ( 7.7 (. ( 7. (. (.70).).0).).0).0).0).).).7).).6).) *Parentheses denote standard deviations tions (Ward and Zahavi 7). Close perching may also aid birds in detecting and avoiding predators (Gadgil 7, Zahavi 7). Or, perhaps numerous branches simply provide abundant, suitable perching positions. The habitat in our autumn roosts is different than that described by most other researchers considering tree roosts (Bliese, Jumber 6, Francis 76, Kelty andlustick 77, Good and Johnson 7) but it is similar to that characterized by Lyon (7). Like us, Lyon (7) examined autumn roosts, while the other workers researched winter aggregations. However, all roosts seem to share the characteristic of a dense canopy. Central Ohio blackbird roosts can be viewed as a middle successional stage and have their beginnings in abandoned crop land, orchards, and pastures. This habitat is a transitional stage and appears to remain attractive to blackbirds for a short period. However, new roost sites emerge on the periphery of expanding cities as farm land is idled through purchase by speculative developers. ACKNOWLEDGMENTS. Salaries and research support provided by state and federal funds appropriated to the Ohio Agricultural Research and Development Center, The Ohio State University. Journal Article No. 6-. We are cooperators in and the study resulted in part from Regional Research Project NE-0. We appreciate the helpful reviews of this paper by Richard Dolbeer and Robert Stehn of the U.S. Fish and Wildlife Service, J. W. Caslick of Cornell University, R.*J. Johnson of University of Nebraska, and R. L. Heiligmann, T. M. Stockdale and C. F. Cole of The Ohio State University. Special thanks go to our research technicians, Bob Schmidt, John Karath, and Randy Smith who put in long hours of field work. Marty Eitel and Dot Micacchion deserve our gratitude for expertly typing the manuscript. LITERATURE CITED Bliese, J. C. W. Species of trees used as roost places by bronzed grackles and avian associates at Ames, Iowa. Proc. Iowa Acad. Sci. 60: 7-. Bloekpoel, H. 76 Bird hazards to aircraft. Books Canada, Inc., Buffalo, NY. 6 p. Cottam, G. and J. T. Curtis 6 The use of distance measures in phytosociological sampling. Ecology 7: -60. Dolbeer, R. A., P. P. Woronecki, A. R. Stickley, Jr., and S. B. White 7 Agricultural impact of a winter population of blackbirds. Wilson Bull. 0: -. Francis, W. J. 76 Micrometeorology of a blackbird roost. J. Wildl. Manage. 0: -6. Gadgil, M. 7 The function of communal roosting: Relevance of mixed roosts. Ibis : -. Good, E.E. 7 Ohio forests, p. -0 In: M. B. Lafferty (ed.), Ohio's natural heritage. Ohio Acad. Sci., Columbus, OH. 6 p.

5 OhioJ. Sci. BLACKBIRD ROOST VEGETATION Good, H. B. and D. M. Johnson 7 Nonlethal blackbird roost control. Pest Control 6: -. Jumber, J. F. 6 ing behavior of the starling in central Pennsylvania. Auk 7: -6. Kelty, M. P. and S. I. Lustick 77 Energetics of the starling (Sturnus vulgaris) in a pine plantation. Ecol. : -. Lyon, L. A. 7 Communal roosting of blackbirds and starlings. Unpubl. M. S. Thesis, Rutgers Univ., New Brunswick, NJ. 6 p. and D. F. Caccamise Habitat selection by roosting blackbirds and starlings: Management implications. J. Wildl. Manage. : -. MacArthur, R. H. and H. S. Horn 6 Foliage profile by vertical measurements. Ecol. 0: 0-0. Meanley, B. 7 Blackbirds and the southern rice crop. U.S. Fish and Wildl. Serv. Resour. Publ p. Mosby, H. S. 7 Reconaissance mapping and map use. p. - In: R. H. Giles (ed.), Wildlife management techniques. The Wildl. Society, Washington, DC. 6 p. Tosh, F. E., I. L. Doto, S. B. Beecher, and T. D. Y. Chin 70 Relationships of starlingblackbird roosts and endemic histoplasmosis. Am. Rev. Resp. Dis. 0: -. Ward, P. and A. Zahavi 7 The importance of certain assemblages of birds as "information centers" for food finding. Ibis : 7. Zahavi, A. 7 The function of pre-roost gatherings and communal roosts. Ibis : 060-