Effect of Forest Clear-Cut Edge on Breeding Birds in East Texas

Breeding birds in 2 pine-hardwood stands (>30 years old) and adjacent clear-cuts ( 30 years old and young pine plantations 30 years old. The clear-cut area was planted in 1975 and 1976. Both areas were ungrazed and had similar site preparations. Bird Census Strips.-We established 4 1 Maintained in cooperation with Stephen F. Austin State University, School of Forestry, Nacogdoches, Texas. J. Wildl. Manage. 44(3):1980 559 This content downloaded from 157.55.39.208 on Fri, 29 Jul 2016 04:50:08 UTC All use subject to http://about.jstor.org/terms 560 HABITAT OF BREEDING BIRDS IN TEXAS * Strelke and Dickson census strips in each study area. Each strip was 80 x 300 m and perpendicular to the woods clear-cut edge. Forty meters either side of the transect center line was sufficient to detect birds effectively. Half of each strip (150 m) was in the clear-cut and half was in the woods. Each census strip was divided into 12 25-m sections. The 2 farthest into the woods were designated Woods Interior (WI), the next 2 Woods Middle (WM), the 2 next to the clear-cut Woods Edge (WE), the Clearcut Edge (CE), the Clear-cut Middle (CM), and the Clear-cut Interior (CI). Bird Census.-Birds in 4 of the 8 census strips were counted on alternate mornings 10 times from 1 May to 23 June 1978 (20 mornings). Counts began at sunrise or shortly thereafter and were completed within 3.5 hours. No counts were taken on mornings with high winds (>32 km/hour) or rain. The censuser walked slowly with occasional pauses along a path bisecting each strip. Each bird visually or aurally detected was recorded on a plot map. Detections with territorial significance, such as singing males, were designated by symbols. Information such as sex, age (mature or immature), mode of feeding, and territorial disputes also was recorded. About 30 minutes was spent on each census strip, half in the woods and half in the clear-cut. Bird-Habitat Association.--Bird species with 10 or more individuals recorded in the censuses were classified by habitat: woods, edge, clear-cut, or all 3 habitats. Species with less than 10 detections were categorized as infrequent observations and treated as a group. Each species was classified in the habitat where >75% of the observations of a species occurred. Vegetation Measurements.-Eight vegetation sample points were randomly located in each census strip: 2 each in woods edge and interior, and in clearcut edge and interior sections. At each point, percentage of ground cover by species, stems per hectare, foliage density by height, and height of dominant and codominant trees nearest the sample point were recorded. Percentage of ground cover was tallied by species for all herbaceous vegetation and woody vegetation 22.9 cm dbh on 405-m2 circular plots. Heights of dominant and codominant trees were measured with an Abney level. Foliage density in the clear-cuts was measured at ground level, 1 m, and 2 m with a 0.41 x 0.41-m grid board (MacArthur and MacArthur 1961). In the woods, foliage density was measured at ground level, 2 m, and 4 m. The grid board was moved horizontally from the observer until half of it was obscured by foliage. Distance from the board to the observer was determined with a range finder to 30 m, and with a tape for distances greater than 30 m. Foliage density in square meters of leaf silhouette per cubic meter of space was calculated at each height as K = (In 2)/D, where In 2 = 0.693, and D = distance from observer to grid board in meters. Foliage density at heights greater than 7 m was measured by sighting vertically through a 400-mm telephoto lens of a reflex camera over each sample point (MacArthur and Horn 1969). Height of vegetation intercepting 16 grid points sketched on the groundglass focusing plate of the camera was measured from the focusing scale of the lens. Foliage density at 7 m and above was calculated from the formula K = ln(pl/p2), where p, = the proportion of measurements above a selected height (h), p2= the proportion of J. Wildl. Manage. 44(3):1980 This content downloaded from 157.55.39.208 on Fri, 29 Jul 2016 04:50:08 UTC All use subject to http://about.jstor.org/terms HABITAT OF BREEDING BIRDS IN TEXAS * Strelke and Dickson 561 measurements above a greater selected height (h2), and K = the number of intercepts in a vertical line between the points. Although theoretically the horizontal and vertical measurements are not identical, in practice they are similar (MacArthur and Horn 1969). Treatment of Data.-Bird species diversity was calculated H' = -1 p, In pi, where pi was the proportion of individuals in the ith species (i = 1, 2, . . . , S). Equitability was calculated as J' = H'/ H'max, where H'max was equal distribution of units (individual birds) among a given number of classes (species). We tested number of bird species, abundance, species diversity, and equitability for differences among 50-m sections and among 25-m subsections by analysis of variance (ANOVA) and Duncan's multiple range test (Duncan's MRT). For an adequate sample of more species we combined data from both study areas. RESULTS AND DISCUSSION Vegetation Characteristics of Study Areas Major ground cover species (>5%) in descending order of area occupied for the Shawnee clear-cut were panic grass (Panicum spp.), shining sumac (Rhus copallina), American beautyberry (Callicarpa americana), blackberry (Rubus spp.), broomsedge (Andropogon virginicus), tree sparkleberry (Vaccinium arboreum), dogfennel eupatorium (Eupatorium capillifolium), and roundseed panicum (Panicum sphaerocarpon). Ground cover on the Cherokee clear-cut was characterized by panic grass, blackberry, shining sumac, pinehill bluestem (Andropogon divergens), and sweetgum (Liquidambar styraciflua). Flowering dogwood (Cornus florida) and longleaf uniola (Uniola sessiliflora) were major ground species in Table 1. Stems and basal area of trees (>22.9 cm dbh) on 2 study areas in East Texas. Basal area Woods Stems/ha (m2/ha)