Dipnet Surveys Provide Accurate Abundance Estimates for Lentic-Breeding Amphibian Larvae.

Several aquatic sampling techniques are commonly used to detect and quantify amphibian larvae. Although many techniques have been standardized, comparisons among methods have been poorly studied. We employed a removal design to investigate the detection rate of amphibian larvae using box enclosures and evaluated the reliability of timed dipnet sampling relative to the more quantitative box enclosure sampling across 13 vernal pools. For these comparisons, we studied larvae of the endangered California Tiger Salamander (Ambystoma californiense; CTS) and the smaller and more abundant tadpoles of the Pacific Treefrog (Pseudacris regilla; PTF). Based on N-mixture model estimates, box enclosure sampling captured only 5% more of the CTS larvae present than PTF tadpoles, despite disparities in their sizes and estimated abundances. Dipnet sampling was slightly more effective at detecting CTS larvae than was enclosure sampling, while both methods equally detected PTF tadpoles. We found a strong relationship (R2 = 0.92) between the densities of CTS larvae estimated from dipnet sampling and those calculated from the more quantitative enclosure sampling, indicating that dipnet sampling is a reliable measurement of relative abundance. We conclude that timed dipnet surveys can be an important tool for monitoring populations of CTS, and possibly other lenticbreeding amphibians, especially given its relative efficiency that can be replicated across both time and space. [ABSTRACT FROM AUTHOR]