Patuxent Wildlife Research Center

NAAMP III Archive - aquatic sampling
Home | Archive by Alphabetical Order | Archive by Category

Larval Amphibian Survey Feasibility for a North American Monitoring Program: Preliminary Findings

Alys Reed

Dept. Environmental Biology
Hood College, Frederick, MD

Due to incongruous and unstandardized information pertaining to amphibian decline, there is an inherent need to define a program which may accurately assess the status of amphibian populations nationwide. Though there are different methods for surveying amphibian populations, this study will specifically consider the prospects of larval surveys for use in a comprehensive amphibian monitoring program

Goals for this Project:

1. Determine what larval survey technique(s) to use for sampling by conducting preliminary test to find which technique(s) (funnel traps, nets, etc.) is most effective for sampling tadpoles;

2. Using the drainable ponds at Patuxent, sample the tadpole population using the appropriate technique(s), drain the ponds and count the entire population;

3. Determine the effectiveness of the sampling technique(s) in terms of how accurately the sampled population depicts the actual population, and formulate an index that can be applied in the field based on this information;

4. Evaluate the feasibility of larval sampling using defined techniques for an amphibian monitoring program.

A series of experimental ponds at Patuxent Environmental Research Center, Laurel, Maryland, were used to investigate larval sampling techniques. Average depth and width of the ponds was 45 cm and 14 m respectively. Ponds were divided into quadrants, each of which was sampled twice; once at a shallow depth and again at a deeper depth. A few ponds were small and of a relatively uniform depth, therefore, only one sample was taken per quadrant. During the period between May and September of 1995 a total of 17 ponds were sampled and drained. Initially both funnel traps and dip nets were used to collect tadpoles. Collection of tadpoles by use of funnel traps yielded only dead individuals, therefore, this technique was eliminated as a reliable and unbiased sampling method. Using 1 meter sweeps with long handled dip nets each pond was sampled a minimum of 24 times. In several instances ponds were sampled by several individuals. Multiple participants helped reduce sampling time. Nonetheless, this factor may have led to the addition of another variable influencing results - observer effect. Ponds were drained either the same day of sampling or on the following day, and all tadpoles were counted.

Within the 17 ponds sampled, 6 different classifications of tadpoles were found. These included small and big leopard frogs (size difference related to position in life cycle), spring peepers, green frogs, cricket frogs, and toads. Variation in species composition amongst ponds sampled can be attributed to differences in breeding time between species. Ponds sampled in May yielded primarily spring peepers and both small and big leopard frogs, whereas ponds sampled in September contained mostly green and cricket frogs.

Results and Discussion:
Preliminary analysis using scatter plot graphs showed a non-linear relationship between total number of tadpoles in each pond and those caught in dip bet samples. Ratios of different species in samples did not correlate with those of the total population. Further considerations revealed that the number of samples taken was inadequate to draw valid conclusions regarding larval surveys. Sampled tadpoles accounted for less than 10% of the total population, subsequently reducing the probability of having obtained a representative sample. Another sampling issue considered during data analysis involved cases where ponds were sampled by alternating between several individuals. The question posed was "what role does the observer play in the overall outcome of the survey?" Possibility of observer effects interfering with outcome could not be addressed because of too many existing unmeasured parameters. Such a test would have required each observer to have been evaluated on technique, consistency, and many other factors as well. In conclusion, this study roughly determined that larval sampling using dip nets does not establish a reliable method for projecting amphibian population size. Further considerations involving sampling issues such as pond profile (depth, vegetation, microhabitats, etc.), and observer effects need to be addressed before dip net sampling is totally discounted as a reliable sampling method.


U.S. Department of the Interior
U.S. Geological Survey
Patuxent Wildlife Research Center
Laurel, MD, USA 20708-4038
Contact: Sam Droege, email:
Last Modified: June 2002