Knowing how many animals are in a population sounds straightforward, but the method you use to count them can dramatically change the answer. Our new study, led by Patrick Moldowan as part of his PhD work at the University of Toronto, draws on 14 years of data from the long-running Bat Lake Integrated Salamander Study / Bat Lake Inventory of Spotted Salamanders (BLISS) at Algonquin Provincial Park to put three common amphibian monitoring approaches head-to-head: aquatic funnel trapping, egg mass counting, and coverboard surveys (Figure 1). By comparing all three against a kilometre-long drift fence encircling the entire lake and intercepted nearly every salamander on its breeding migration, we were able to ask which methods actually give reliable estimates of population size and sex ratio for spotted salamanders (Ambystoma maculatum). The drift fence revealed annual breeding populations of between 4,500 and 5,300 individuals, a strikingly large population that reflects the intact, fish-free nature of Bat Lake.

The results carry some important cautionary lessons for wildlife monitoring. Funnel traps alone underestimated population size by 67–75%, largely because individuals differ enormously in how likely they are to be caught. When this individual heterogeneity in capture probability was accounted for in the models, estimates aligned much more closely with the fence counts. Even more striking was what the traps suggested about sex ratios: raw trap data implied males outnumbered females two-to-four-to-one, when in reality the population was slightly female-biased. Coverboards, meanwhile, detected less than 1% of the population and were essentially uninformative for this species. Egg mass counts, often used as a convenient proxy for female abundance, were only weakly correlated with actual female numbers at this site, showing that these monitoring shortcuts need to be validated before being relied upon. Together, these findings offer practical guidance for anyone designing long-term amphibian monitoring programs, emphasizing that how you count is just as important as what you count.

Citation

Moldowan, PD, Armstrong, DP, Tattersall, GJ, and Rollinson, N. 2026. A comparison of monitoring methods for inferring demographics of a pond-breeding amphibian population over 14 years. Journal of Wildlife Management, e70225. https://doi.org/10.1002/jwmg.70225