Relative Abundance of Day-Flying Insects in Treefall Gaps vs Shaded Understory in a Neotropical Forest

Abstract

Day-flying insects were sampled in treefall gaps and closed understory within the semideciduous, lowland forest of Barro Colorado Island, Panama, from April (late dry season) to August (mid wet season) in 1980. Transparent strips of plastic coated with tanglefoot were suspended above ground at 5 treefall and 5 understory sites. Traps were run from early morning to late afternoon for 4-day periods at approximately 2-wk intervals. For most taxa, the numbers of individuals captured were relatively low during the first sampling period (prior to the onset of heavy rains), increased to peak levels between mid-May and mid-June, and then declined slightly in July and August. Statistically significant differences in abundance between gap and understory sites were noted only for Formicidae, Coleoptera, and Psocoptera. In all 3 cases, greater numbers of individuals were trapped at understory sites. Size frequency distributions were similar between treefall gaps and understory for all taxa. A PROMINENT FEATURE OF MANY TROPICAL FORESTS iS structural heterogeneity resulting from the occurrence of treefall gaps. Gaps of varying sizes and ages frequently interrupt the shaded understory and create conspicuously distinct habitat patches within many tropical forests (Halle et al. 1978, Whitmore 1978, Lang & Knight 1983). Vegetational differences between treefall gaps and the closed understory frequently have been reported (Oldeman 1978, Denslow 1980), and the role of treefall gaps in tropical forest dynamics has received considerable attention recently (Hartshorn 1980, Foster & Brokaw 1982, Brokaw 1985). In contrast, the influence of treefall gaps upon the structure of animal communities is poorly understood. Schemske and Brokaw (1981) censused birds in treefall gaps and adjacent forest in central Panama and found pronounced differenes in species richness and composition. Working in the same region, I described distinct lightand shadeseeking behavior of adult damselflies (Shelly 1982) and robber flies (Shelly 1984, 1985), both sit-and-wait predators that attempt aerial capture of flying insects. For both taxa, certain species were found to perch and forage primarily within the shaded understory, whereas others were found almost exclusively in light gaps. Recognition of these lightand shade-seeking guilds, coupled with additional behavioral and ecological observations, raised questions regarding possible differences in the spectra of available prey between treefall gaps and closed understory. The great majority of studies involving large-scale sampling of insects within tropical habitats does not deal with the day-flying, aerial fauna. Instead, most prior studies sample insects in one of these modes: attracted to lights at night (Wolda 1978, Wolda & Fisk 1981); resting on vegetation by sweeping, beating, or fogging (Janzen 1973, Janzen & Pond 1975); occurring in leaf litt r or in the soil (Levings & Windsor 1985); attracted to mammalian baits (Goodier 1966, Clarke 1968); attracted to rotting fruit (Dobzhansky & Pavan 1950, Owen & Chanter 1972); or attracted to pheromone baits (Ackerman 1983, Greenfield 1983). Malaise traps have been used successfully in some studies (Owen & Chanter 1970, Denlinger 1980), but the collections may have been highly site-dependent, since only one or two traps were operated in fixed locations. Additionally, of course, Malaise traps capture both dayand night-flying insects. Other methods previously used to sample day-flying insects in tropical habitats include tow-netting (Hespenheide 1975), yellow water traps (Duviard & Pollett 1973), window pane traps (Robinson & Robinson 1973), sticky traps (Robinson & Robinson 1973), and suction traps (Bigger & Tapley 1969, Rose 1972). However, in none of these studies (or those using Malaise traps) was an attempt made to systematically sample and compare treefall gaps and understory within primary forest. Moreover, in a number of these studies (Bigger & Tapley 1969, Owen & Chanter 1970, Rose 1972), samples were taken in urban or agricultural areas. The present study provides the results of a sampling program designed to compare the relative abundance of day-flying insects in treefall gaps vs shaded understory in a lowland Neotropical forest. Data regarding seasonal trends in abundance and size distributions are presented as well. Insects were sampled using transparent, tanglefoot-covered traps suspended above the ground. Although this method unquestionably yields biased samples (discussed below), the use of tanglefoot traps was considered justified for five reasons: their low cost (in both dollars and labor) of construction, shipping, and maintenance; their light weight and portability; their presumably nonattractive nature that allows collection of site-specific samples; their ease of opI Received 31 March 1986, revision accepted 11 September 1986. 114 BIOTROPICA 20(2): 114-119 1988 This content downloaded from 157.55.39.35 on Fri, 02 Sep 2016 05:39:13 UTC All use subject to http://about.jstor.org/terms eration by a single person; and the reproducibility of the sampling protocol. Additionally, although canopy insects may be abundant within tropical forests (Sutton & Hudson 1980), sampling in the present study was confined to the ground level for logistic reasons.