Do frugivores respond to fruit harvest? An experimental study of short-term responses

Many nonexperimental studies have shown positive correlations between resource abundance and vertebrate abundance. These relationships, however, are difficult to interpret because of confounding factors that may independently determine the abundance of vertebrates and their resources. Moreover, verifying links between vertebrate and resource abundance is complex because human perception of resource abundance may differ from that of nonhuman vertebrates and because manipulating resource abundance on an ecologically meaningful scale is difficult. We studied the dependency of frugivores on fruit abundance in eastern Amazonian floodplain forests dominated by one species of palm tree (Euterpe oleracea) from which people harvest fruit. We first compared spatial and temporal use by frugivorous parrots of four sites dominated by E. oleracea and four sites with no E. oleracea. Parrots spent 48-92% more time in the former, where their activity over the fruiting season mirrored the abundance of fruits. To test whether fruit abundance was the mechanism underlying these patterns, we removed fruit at two intensities in replicated 1.8-ha plots, and then monitored responses of frugivorous birds and mammals. High-intensity removal (75% of ripe fruit harvested) significantly reduced the number of frugivorous bird individuals by 29% and the length of frugivorous bird visits by 68%, relative to controls. In contrast, low-intensity removal (41% of ripe fruit removed) had no impact on these metrics. Frugivore species richness did not differ among treatments and controls, but the composition of the frugivore community was altered by harvest, with the presence of 11 species being linked to fruit abundance. Nonfrugivorous birds did not respond to either intensity of fruit harvest. The number of fruit-eating mammal species was 58% lower in both the low- and high-removal treatments, relative to control plots. These results verify that fruit abundance influences the species composition of frugivore communities and the abundance and foraging behavior of individual species. They also document short-term dependency of fruit-eating mammals on fruit abundance. Harvest of fruit from forest systems, a common practice in tropical forests, therefore can affect populations of fruit-eating animals.