Bird species diversity on an andean elev ational gradient

This paper analyzes patterns of bird species diversity on an elevational transect of the Cordillera Vileabamba, Peru. Major changes in climate and vegetation are encompassed by the transect which extended from the Apurimac Valley floor at 500 m to the summit ridge of the range at > 3,500 m. Four vegetation zones are easily discerned-lowland rain forest, montane rain forest, cloud forest, and elfin forest. In progressing upwards there is a monotonic trend toward decreasing canopy stature and reduced number of plant strata. The vegetation gradient provided the opportunity to examine the relation between bird species diversity and habitat complexity in an entirely natural setting. The decrease in forest stature with elevation was closely paralleled by decreasing avian syntopy (the total number of bird species cohabiting the forest at a given elevation). Bird species diversity was shown to be highly correlated with foliage height diversity, using either four or five layers in the foliage height diversity calculation (r = .97), and less well correlated using three layers, as defined previously by MacArthur (r = .84). At this superficial level the trend in bird species diversity seemed to be adequately explained as a response to the vegetation gradient. This preliminary conclusion was found to be illusory when the elevational trend in syntopy was reexamined separately for three major trophic subdivisions of the fauna. The number of insectivores decreased 5.2-fold from the bottom to the top of the gradient, frugivores decreased by a factor of 2.3, and nectarivores showed no change. It was now clear that the diversity in each of these trophic categories was responsive to environmental influences other than, or in addition to, the gradient in habitat structure. Additional factors implicated by the available evidence are competitive interactions with other taxa at the same trophic level, changing composition of the resource base as a function of elevation, and declining productivity at high elevations. Analysis of netted bird samples revealed an unexpected diversity maximum in the lower cloud forest zone. The immediate cause of this was a relaxation of the vertical stratification of foraging zones, such that an anomalously large fraction of the species present entered the nets. The excess diversity was found to consist almost entirely of insectivores. Several factors appear to contribute to the ultimate causes of the diversity maximum: greater patchiness of the montane forest due to the rugged topography, a higher density of foliage near the ground, and possibly increased resource productivity. A correlation between diversity and density in the netting results suggested a causal connection mediated via resource levels. The conclusion is that diversity is a complex community property that is responsive to many types of influences beyond simply the structure of the habitat.

Duke Scholars

Author John W. Terborgh Environmental Sciences and Policy