Detection probabilities of two introduced frogs in Hawaii: implications for assessing non-native species distributions

Two nonnative Caribbean frogs, the Puerto Rican coqui and the Cuban greenhouse frog, recently invaded Hawaii. Because of its louder breeding call, management efforts have focused on the coqui, while little has been done to address the more cryptic greenhouse frog, even though it may be as widespread and have similar ecological impacts. The goal of this research was to determine the distribution and detection probability of both species on the island of Hawaii. We conducted a breeding call presence/absence survey at 446 sites every 2 km along major road networks. We re-surveyed 125 sites twice to determine detection and occupancy probabilities. Greenhouse frog detection probabilities (0.24, 0.29, 0.48, for each of the three visits, respectively) were lower than coqui detection probabilities (0.58, 0.73, 0.50, respectively) and increased with visits while those of the coqui did not. Greenhouse frog detection probabilities were lower in the presence of coquis for the first two surveys (0.12, 0.14) than in sites with greenhouse frogs alone (0.41), while greenhouse frogs had no effect on the detection of coquis. Site occupancy estimates for the greenhouse and coqui frog were 0.35 and 0.31, respectively, suggesting the species are similarly widespread. Results suggest multiple visits to sites are required to detect the greenhouse frog. Furthermore, results suggest that accounting for detectability is essential when determining the extent of invasion of cryptic species.