Björn Lardner

Movements and Activity of Juvenile Brown Treesnakes (Boiga irregularis)

Understanding the spatial ecology and foraging strategy of invasive animals is essential for success in control or eradication. We studied movements and activity in juvenile Brown Treesnakes on Guam, as this population segment has proven particularly difficult to control. Distance between daytime refugia (from telemetry of 18 juveniles, 423–800 mm snout–vent length) ranged from 0–118 m (n  =  86), with a grand mean of 43 m. There were tendencies for shorter snake movements on nights directly following a full moon and on dry nights, but variation among snakes was of a larger magnitude and would greatly reduce chances to detect moon or rain effects unless corrected for. Snake activity was estimated from audio recordings of signals from “tipping” radio transmitters, analyzed for pulse period and amplitude. Activity was highest in the hours immediately after sunset, and gradually declined throughout the night before dropping abruptly in conjunction with sunrise. Snake activity was higher on rainy nights, and tended to be highest during waning moons and when the moon was below the horizon. We conclude that small Brown Treesnakes forage actively and appear to move far enough to regularly encounter the traps and bait used on Guam for control purposes, suggesting that alternative explanations are required for their low capture rates with these control tools.

The results of nocturnal visual surveys are influenced by lamp properties

We conducted standardized visual searches at night for brown treesnakes (Boiga irregularis) and geckos, where we alternated between spotlight and floodlight lamps. Floodlights rendered us 25% more snakes and 71% more geckos than did spotlights. We show data on searcher variability and discuss what might affect the relative benefit of different lamp types.

Detection Rates of Geckos in Visual Surveys: Turning Confounding Variables into Useful Knowledge

Abstract Transect surveys without some means of estimating detection probabilities generate population size indices prone to bias because survey conditions differ in time and space. Knowing what causes such bias can help guide the collection of relevant survey covariates, correct the survey data, anticipate situations where bias might be unacceptably large, and elucidate the ecology of target species. We used negative binomial regression to evaluate confounding variables for gecko (primarily Hemidactylus frenatus and Lepidodactylus lugubris) counts on 220-m-long transects surveyed at night, primarily for snakes, on 9,475 occasions. Searchers differed in gecko detection rates by up to a factor of six. The worst and best headlamps differed by a factor of at least two. Strong winds had a negative effect potentially as large as those of searchers or headlamps. More geckos were seen during wet weather conditions, but the effect size was small. Compared with a detection nadir during waxing gibbous (nearly full) moons above the horizon, we saw 28% more geckos during waning crescent moons below the horizon. A sine function suggested that we saw 24% more geckos at the end of the wet season than at the end of the dry season. Fluctuations on a longer timescale also were verified. Disturbingly, corrected data exhibited strong short-term fluctuations that covariates apparently failed to capture. Although some biases can be addressed with measured covariates, others will be difficult to eliminate as a significant source of error in long-term monitoring programs.

Stability of Detectability over 17 Years at a Single Site and other Lizard Detection Comparisons from Guam

Abstract To obtain quantitative information about population dynamics from counts of animals, the per capita detectabilities of each species must remain constant over the course of monitoring. We characterized lizard detection constancy for four species over 17 yr from a single site in northern Guam, a relatively benign situation because detection was relatively easy and we were able to hold constant the site, habitat type, species, season, and sampling method. We monitored two species of diurnal terrestrial skinks (Carlia ailanpalai [Curious Skink], Emoia caeruleocauda [Pacific Bluetailed Skink]) using glueboards placed on the ground in the shade for 3 h on rainless mornings, yielding 10,286 skink captures. We additionally monitored two species of nocturnal arboreal geckos (Hemidactylus frenatus [Common House Gecko]; Lepidodactylus lugubris [Mourning Gecko]) on the basis of 15,212 sightings. We compared these count samples to a series of complete censuses we conducted from four or more total removal plots (everything removed to mineral soil) totaling 400 m2 (about 1% of study site) in each of the years 1995, 1999, and 2012, providing time-stamped quantification of detectability for each species. Unfortunately, the actual population trajectories taken by the four species were masked by unexplained variation in detectability. This observation of debilitating latent variability in lizard detectability under nearly ideal conditions undercuts our trust in population estimation techniques that fail to quantify venue-specific detectability, rely on pooled detection probability estimates, or assume that modulation in predefined environmental covariates suffices for estimating detectability.

Inferring the absence of an incipient population during a rapid response for an invasive species

Successful eradication of invasives is facilitated by early detection and prompt onset of control. However, realizing or verifying that a colonization has occurred is difficult for cryptic species especially at low population densities. Responding to the capture or unconfirmed sighting of a cryptic invasive species, and the associated effort to determine if it indicates an incipient (small, localized) population or merely a lone colonizer, is costly and cannot continue indefinitely. However, insufficient detection effort risks erroneously concluding the species is not present, allowing the population to increase in size and expand its range. Evidence for an incipient population requires detection of ≥1 individual; its absence, on the other hand, must be inferred probabilistically. We use an actual rapid response incident and species-specific detection estimates tied to a known density to calculate the amount of effort (with non-sequential detections) necessary to assert, with a pre-defined confidence, that invasive brown treesnakes are absent from the search area under a wide range of hypothetical population densities. We illustrate that the amount of effort necessary to declare that a species is absent is substantial and increases with decreased individual detection probability, decreased density, and increased level of desired confidence about its absence. Such survey investment would be justified where the cost savings due to early detection are large. Our Poisson-based model application will allow managers to make informed decisions about how long to continue detection efforts, should no additional detections occur, and suggests that effort to do so is significantly higher than previously thought. While our model application informs how long to search to infer absence of an incipient population of brown treesnakes, the approach is sufficiently general to apply to other invasive species if density-dependent detection estimates are known or reliable surrogate estimates are available.