Editorial Type:
Article Category: Research Article
 | 
Online Publication Date: 30 Dec 2013

Habitat Selection by Bothrops atrox (Serpentes: Viperidae) in Central Amazonia, Brazil

,
,
,
, and
Page Range: 684 – 690
DOI: 10.1643/CE-11-098
Save
Download PDF

Tropical rainforests often appear relatively homogeneous on satellite images, but responses to landscape characteristics may be found on finer scales if habitat characteristics are considered as continuous variables. In this study, we used 30 uniformly distributed plots and 16 plots beside streams to evaluate the effects of distance from stream, litter depth, altitude, slope, and tree density on abundance of Amazonian Lancehead Pitviper (Bothrops atrox). We estimated densities and probabilities of detection of snakes in riparian and upland plots in Reserva Ducke, Manaus, Amazonas, Brazil. Apparent density of individuals of B. atrox was about 6.4 times higher near streams, but the number of individuals in the landscape more than 10 m away from streams was about 3.9 times higher than the number of individuals within 10 m of streams. Movement data from two adult B. atrox evaluated by radio telemetry indicate that individuals can disperse out of plots and away from streams over a period of several months. Detectability of B. atrox varied little among riparian and upland plots, so differences in detectability are unlikely to be responsible for large differences in encounter rates of snakes between riparian and non-riparian areas. There were small differences in body size of individuals near streams and individuals far from streams. The distribution of B. atrox is not uniform within the forest. However, as with most other tropical-forest organisms studied to date, this species occurs across wide environmental gradients and shows only subtle habitat specificity.

Copyright: 2013 by the American Society of Ichthyologists and Herpetologists
Fig. 1. 
Fig. 1. 

Location of Reserva Ducke, Manaus, Amazonas, Brazil, highlighting the 25 km2 RAPELD grid. Circles represent uniformly distributed plots, black squares represent riparian plots, and open squares indicate uniformly distributed plots located close to streams.


Fig. 2. 
Fig. 2. 

Relationships between snout–vent length of individuals of Bothrops atrox (A) and number of individuals per plot (B), and log10 distance from the streams. Black circles  =  one plot, squares  =  two overlapped plots, triangle  =  three overlapped plots, white circle  =  four overlapped plots.


Fig. 3. 
Fig. 3. 

Detection distance of individuals of Bothrops atrox per habitat. R  =  riparian plots, U  =  uniformly distributed plots.


Fig. 4. 
Fig. 4. 

Positions of the two radio-tracked individuals of Bothrops atrox in relation to distance along and from the streams. Dotted lines  =  strip 10 m wide each side of the stream, circles  =  positions in July, triangles  =  positions in October–November.


Contributor Notes

Associate Editor: J. D. Litzgus.

Received: 15 Jul 2011
Accepted: 07 May 2013
  • Download PDF