Observing this relative decline in amphibian populations has elicited grave concern on the part of herpetologists and ecologists (Pechmann et al. 1991, Pechmann and Wilbur 1994, Pechmann and Wake 1997). However, an important question remains - whether these declines are simply the product of natural fluctuations in species abundance or a direct result of human interference (Gibbons et al. 1997, Pechman et al. 1991)? To properly assess anuran populations and the factors affecting them, detailed monitoring surveys are necessary. Unfortunately, such accurate appraisals of amphibian populations in most areas have not been achieved.
Today, many amphibian monitoring programs rely on manual calling surveys to evaluate anuran populations. Certain benefits accompany manual surveys of anuran populations, most notably the extensive site coverage (e.g. data collection from multiple sites) they offer. However, there are several problems often associated with manual calling surveys. First, manual surveys do not allow continuous sampling. Secondly, manual calling surveys often lack standard techniques or universally accepted conditions and do not provide permanent sampling records. In addition, manual surveys do not account for observer biases; only the investigator is charged with identifying anuran species and quantifying their numbers. Finally, and perhaps most importantly, manual calling surveys can interfere with the natural behavior of the species being surveyed.
The use of automated recording systems can overcome many of the limitations of traditional manual sampling techniques. Although such surveys do not utilize direct on-site observation, automated sampling has several advantages when compared with manual sampling procedures (Peterson and Dorcas 1994). With automated recording surveys, 1) continuous sampling is possible, 2) a permanent sampling record is available, allowing multiple investigators to evaluate calls, 3) the observer is free to complete other projects, and 4) disturbance of animal activity is minimized.
As discussed by Peterson and Dorcas (1992), variation in environmental conditions greatly affects animal activity. Especially in ectotherms such as amphibians, changes in the environment closely correlate with changes in amphibian activity (Peterson and Dorcas 1992). When combined with environmental data, information gained from automated recording systems can be used to determine the effects of environmental variation on anuran calling activity. The ability to predict animal activity, and thus visibility, is vital to the optimization of manual monitoring programs.
This research aims to develop a model predicting anuran calling activity based on environmental variation. To achieve this goal, the following research must be performed: 1) an accurate assessment of anuran calling activity in a particular area, 2) a simultaneous, detailed record of environmental variation in the area, and 3) the combination of these results to determine how environmental variation affects anuran calling behavior. By formulating a model that accounts for various environmental factors affecting amphibian vocalization, the probability of detecting anurans using manual surveys in a specific area can be calculated given certain environmental measurements. Since a lack of vocalization does not necessarily imply the absence of amphibians, such a model would also allow correction of data collected in manual calling surveys.
