UCLA

The Maya Forest Monitoring Project (Mayamon) was established in 1997 as an outgrowth of the Belize working group of the Declining Amphibian Populations Task Force. For several years, Mayamon volunteers have censused anuran populations using a protocol that estimates numbers of individuals on the basis of male vocalization. To date, the protocol has been evaluated only through a series of post hoc power analyses; I performed the first field test to assess the effect of species-specific mating system characteristics, survey length, survey frequency, and pond selection on census results for anuran communities within a tropical moist forest in Belize. Under the current protocol, I estimate that it would take 359 months of sampling to detect the 11 species I detected at this site using vocalization surveys. In addition, I introduce a method using ANCOVA to determine ideal survey length. Arbitrarily setting the ideal detection to 90% yields a required sampling protocol of 21 minutes; the current minimum of 15 minutes yields only an 80% detection rate. This method could be adapted for use with other monitoring programs, allowing both the assessment of current efficacy and the extrapolation of required sampling length to reach a given efficacy. The results of these approaches indicate that the Mayamon protocol methodology should be extended if it is to allow investigators to adequately understand the community dynamics of amphibians.