Editorial Type:
Article Category: Research Article
 | 
Online Publication Date: 11 Jun 2021

The Use of Evoked Vocal Responses to Detect Cryptic, Low-Density Frogs in the Field

,
,
,
,
, and
Page Range: 174 – 180
DOI: 10.1670/19-079
Save
Download PDF

ABSTRACT

The protection of rare amphibians depends upon accurate information about their distributions, yet cryptic, low-density populations are easily missed during field surveys. We used sites with known populations of the U.S. federally threatened species Rana chiricahuensis (n = 34) and its congener Rana yavapaiensis (n =11) in Arizona, USA to test whether a conspecific call stimulus can improve detectability of species. We show that the use of the evoked vocal responses technique improves the detectability of initially silent anuran populations of R. chiricahuensis and R. yavapaiensis by 45% and 71%, respectively. Evoked vocal responses, however, were not strictly species specific: 19% of R. chiricahuensis and 45% of R. yavapaiensis populations also increased their vocal activity in response to heterospecific stimuli. The low costs of the material (<50 USD per set) and the short time (1 min) to play a stimulus suggest that the additional costs associated with this technique will be minor relative to existing survey protocols. This technique offers the potential to improve the detection of other frog species, particularly in sites that are difficult to search visually, provided it is carried out by surveyors who can identify species by their calls.

Copyright: Copyright 2021 Society for the Study of Amphibians and Reptiles 2021
Fig. 1.
Fig. 1.

Summary of calling responses of R. chiricahuensis focal populations to a conspecific stimulus. Surface area of circles are proportional to sample sizes.


Fig. 2.
Fig. 2.

Summary of calling responses of R. yavapaiensis focal populations to a conspecific stimulus. Surface area of circles are proportional to sample sizes.


Fig. 3.
Fig. 3.

Plots of population calling behavior and environmental covariates in R. chiricahuensis. Sampled populations (n = 34) from Figure 1 are assigned to one of three categories: initially calling before audio stimulus (n = 3), started calling in response to a conspecific audio stimulus (n = 14), and remained silent after conspecific stimulus (n = 17). A slight statistical jitter (±10%) is intentionally added to wind speed and months of year to avoid overlapping points.


Contributor Notes

Corresponding Author. E-mail: martin.schlaepfer@unige.ch

Current Address: 1335 North Dusty Hollow Court, Tucson, Arizona, 85745, USA

Accepted: 15 Jan 2021
  • Download PDF