Using visual (VI) and dip net indices (DI) and double-observer (DOE), removal (RE), and neutral red dye capture-recapture (CRE) estimates, we counted, estimated, and censused Couch's spadefoot (Scaphiopus couchii) and canyon treefrog (Hyla arenicolor) tadpole populations in Big Bend National Park, Texas. Initial dye experiments helped us determine appropriate dye concentrations and exposure times to use in mesocosm and field trials. The mesocosm study revealed higher tadpole detection rates, more accurate population estimates, and lower coefficients of variation among pools compared to those from the field study. In both mesocosm and field studies, CRE was the best method for estimating tadpole populations, followed by DOE and RE. In the field, RE, DI, and VI often underestimated populations in pools with higher tadpole numbers. DI improved with increased sampling. Larger pools supported larger tadpole populations, and tadpole detection rates in general decreased with increasing pool volume and surface area. Hence, pool size influenced bias in tadpole sampling. Across all techniques, tadpole detection rates differed among pools, indicating that sampling bias was inherent and techniques did not consistently sample the same proportion of tadpoles in each pool Estimating bias (i.e, calculating detection rates) therefore was essential in assessing tadpole abundance. Unlike VI and DOE, DI, RE, and CRE could be used in turbid waters in which tadpoles are not visible. The tadpole population estimates we used accommodated differences in detection probabilities in simple desert pool environments but may not work in more complex habitats.
Additional Publication Details
An evaluation of population index and estimation techniques for tadpoles in desert pools