Appendix A. Supplementary methods for literature search, study selection, effect size calculations, and analyses of bird predation and diversity effects.
Literature search
We searched the literature for publications on the quantitative effects of vertebrate insectivores, using the ISI Web of Science database. We also sent out a request for unpublished data on the Ecolog-Listserv.
Study selection
Often, one publication reported separate experimental comparisons in qualitatively different sites or in different forest layers. Following conventional meta-analytical reviews (Gurevitch et al. 1992, Halaj and Wise 2001), we considered these comparisons as separate studies. For multi-year studies using the same exclosures over the same plants, we used values from the final year of study.
Effect size calculations
For studies which did not report by all arthropods, herbivores or carnivores, we summed the means of the appropriate taxonomic groups or guilds to create these variables. When studies reported repeated measures, we calculated effect sizes for overall means and endpoint data. We observed that the trends were similar with either calculation so we used endpoints for reporting and the forest vs. agroforest analysis.
Analyses of bird diversity and predation effects
We used data collected from exclosure studies to calculate the percent reduction in total and large (>5 mm) arthropods between control and exclosure treatments. We used data from 10 min 25-m radius point counts available from the same habitat types in study regions in which nine bird exclosure studies (by crop, eight in coffee, one in cacao; by strata, three in the canopy, and six in crop plants) were conducted. Nearly all of the point counts were conducted by the same two observers (PB and RSG) thus lessening observer bias. RSG categorized all birds recorded in each study into guilds (frugivore, nectarivore, granivore, insectivore, and omnivore), primary foraging strata (canopy, crop layer, ground), and migratory status (migrant or resident). Changes in the behaviors and activities of resident birds in the breeding season may significantly alter their effects on arthropods (Greenberg 1995) and our complete data set included very few studies that spanned both the dry and wet season. We thus created three bird variables per site; all insectivorous (including omnivorous) birds, migrants, and residents. Once bird data sets were organized, we created sample- and individual-based rarefaction curves with MaoTao calculations in EstimateS 7.5 (Colwell and Coddington 1994; http://viceroy.eeb.uconn.edu/estimates, Colwell et al. 2004). For sample-based rarefaction, we examined bird abundance and species density for all, migrant, and resident birds seen in 38 points – the no. of samples in study sites with the fewest points during the dry season. For individual-based rarefaction, we examined the number of birds found for the total no. of individual birds (132), migrants (65), and residents (29) seen in the site with the lowest bird abundance for each group. We chose to look at both sample- and individual-based rarefaction in order to examine influences of (a) bird density (including the number of species and individuals for a set number of samples) and (b) species richness (species found per a set number of individuals) on bird predation. To examine relationships between bird assemblages and bird effects we performed linear regressions using percent reduction of all and large (>5mm) arthropods as the dependent variables and no. of total, migrant, and resident individuals per sample (density of individuals), no. of total, migrant, and resident species per sample (species density) and per individual (species richness).
LITERATURE CITED
Colwell, R. K., and J. A. Coddington. 1994. Estimating terrestrial biodiversity through extrapolation. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 345:101–118.
Colwell, R. K., C. X. Mao, and J. Chang. 2004. Interpolating, extrapolating, and comparing incidence-based species accumulation curves. Ecology 85:2717–2727.
Greenberg, R. 1995. Insectivorous migratory birds in tropical ecosystems: the breeding currency hypothesis. Journal of Avian Biology 26:260–263.
Gurevitch, J., L. L. Morrow, A. Wallace, and J. S. Walsh. 1992. A meta-analysis of competition in field experiments. American Naturalist 140:539–572.
Halaj, J., and D. H. Wise. 2001. Terrestrial trophic cascades: How much do they trickle? American Naturalist 157:262–281.