FOOD FOR EAmY SUCCESSION BIRDS: RELATIONSHIPS AMONG ARTHROPODS, SHRUB VEGETATION, AND SOIL

-During spring and early summer, shruband herbaceous-level vegetation provides nesting and foraging habitat for many shrub-habitat birds. We examined relationships among arthropod biomass and abundance, foliage leaf surface area and weight, vegetation ground cover, soil characteristics, relative humidity, and temperature to evaluate what factors may influence arthropod food resources for birds. Relative humidity was inversely associated with arthropod biomass; as humidity increased biomass decreased (r = -0.44, P = 0.004). We failed to detect any relationships between deciduous foliage (surface area and weight) and arthropod biomass or abundance. However, both arthropod abundance (r = 0.30, P = 0.06) and biomass (r = 0.39, P = 0.01) were positively associated with the percentage of herbaceous ground cover. Arthropod abundance also appeared to be positively associated with the percentage of clay in the soil and negatively associated with the percentage of sand. Herbaceous layer vegetation (forbs and grasses) is known to be positively associated with fire frequency suggesting a possible foraging benefit for birds during spring in habitats that are frequently burned. Management of early and late succession pine forest habitat to produce and maintain a healthy herbaceous layer will likely support more arthropods and provide quality foraging habitat for birds. Many foliage-gleaning birds are dependent on arthropods as food, particularly during spring and early summer (Berthold 1976, Lewke 1982). In late summer and fall, bird use of plant materials such as fruits generally increases (Baird 1980). Many species of birds breed in early successional habitats created by timber harvesting (Conner and Adkisson 1975, Dickson et al. 1995). Such habitat typically has an abundance of both herbaceous and shrub-level vegetation. Recent research suggests that herbaceous ground cover, which is associated with frequent fire in the south, may be more important for the production of arthropod biomass and abundance than shrub leaf surface area (Hess and James 1998, James et al. 2001, Collins et al. 2002). We explored the relationships among soil characteristics, vegetation, and arthropods in a 3-year-old pine plantation with deciduous and pine foliage that was present in early successional vegetation in Nacogdoches County, eastern Texas, during June 1982. If soil nutrients were a determinant of leaf nutritional quality, sites with better soil nutrients should produce higher arthropod abundance and biomass. We examine possible relationships between weather, vegetation, and soil characteristics and arthropod biomass and abundance. STUDY SITE We selected a 3-year-old loblolly pine (Pinus taeda) plantation (20 ha) with pine, deciduous, and herbaceous foliage from ground level to about l .5 m high on the Angelina National Forest (3 lo 15' N, 94" 15' W) in eastern Texas. The plantation had patchy foliage and ranged from xeric, sandy hilltops to mesic sites along intermittent streams. Loblolly pine, shortleaf pine (Pinus echinata), post oak (Quercus stellafa), winged sumac (Rhus copallina), smooth sumac (R. glabra), and sweetgum (Liquidambar styraciflua) were the dominant woody plants in the young pine plantation. METHODS Arthropod biomass and abundance were estimated by sampling 40 rectangular volumes of foliage (1 x 1 x 10 m) with a 38-cm-diameter insect sweep net. The sampled rectangular volumes (our sample unit) were 'E-mail: [email protected] Bull. Texas Omith. Soc. 39(1): 2006 at least 20 m apart to avoid violating independence of observation. Twenty net sweeps were made of the foliage within each of these rectangular volumes between 0830 and 1130 DST from 22 June to 21 July 1982, and captured arthropods were placed in a kill-jar with chloroform. Arthropods in these samples were identified to taxonomic order and oven dried at 85' C for 48 h (to constant weight) and weighed on an analytical balance. Relative humidity was measured with a sling psychrometer and ambient air temperature recorded immediately after arthropods were sampled. Percent ground cover of herbaceous dicots (forbs) and monocots (grasses) was estimated by viewing downward through a 4 cm diameter x 13 cm long hollow PVC tube at three points systematically placed on the 10-m-long rectangle. A soil auger was used to collect a soil Sample to a depth of 15 cm for each rectangle at the three points where ground cover was estimated and the Samples from each of these three points combined and mixed in a soil sample bag. Soil'samples were analyzed for pH, texture, and chemical properties at the Stephen F. Austin State University Department of Agriculture Soil Science Laboratory. All foliage within each of the 1 x 1 x 10 m rectangular volume (n = 40) where arthropods had been sampled was identified to species, clipped with hand pruning shears, and placed in a separate, large plastic bag for each rectangle and transported back to the laboratory. In the laboratory each leaf was run through a Licor area meter (LI-3000) with a conveyor belt three times, and then averaged to measure total leaf surface area for each rectangular volume. Foliage from each sample unit was then oven dried to constant weight at 85" C to obtain a measure of foliar biomass for each rectangular volume. Foliar surface area and dry weight biomass were divided by 10 to obtain a measure per cubic meter. All variables were evaluated for normality (Kolmogorov-Smirnov one sample test, P < 0.05) and transformed with an arcsine transformation, if necessary, and descriptive statistics were calculated (Table 1). Relationships among variables were examined using Pearson correlations. Multiple linear regressions (forward stepwise) were used to determine what subsets of variables were the best predictors of arthropod biomass and abundance. Table 1. Descriptive statistics for arthropod, foliage, micro climate, and soil characteristics in early succession shrub-level vegetation (n = 40) in eastern Texas. Habitat characteristic Mean SD Minimum Maximum Arthropod biomass (g) 0.09 0.09 0.003 0.386 Arthropod abundance (#) 18.7 12.2 2.0 55.0 Relative humidity (%) 82.7 9.3 58.0 95.0 Temperature (CO) 29.0 3.2 21.5 35.3 Foliage leaf area (cm2/m3) 1415.8 538.7 548.7 3310.3 Foliage weight (g/m3) 143.1 47.9 66.8 281.1 Herbaceous dicot (%) 36.3 17.0 7.0 75.0 Herbaceous monocot (8) 51.2 16.5 10.0 78.0 Sand (%) 73.9 2.9 68.0 80.0 Silt (%) 11.0 2.2 6.0 16.0 clay (%I 15.2 1.8 12.0 22.0 Soil pH 5.9 0.2 5.3 6.3 Calcium (ppm) 717.5 229.7 300.