Sampling Reptiles in the Anthropocene - via Herp Digest

By C. Kenneth Dodd 6/30/16, Oxford University Blog

According to the Reptile Database, more than 10,200 non-avian reptile species have been described (6,175 lizards and amphisbaenians; 3,496 snakes; 341 turtles; 25 crocodilians; 1 tuatara), with new taxa being recognized nearly every day. Reptiles are not on the mind of most people in the so-called “developed” world, but they play a major role in the ecology of many regions as predators and prey, particularly in tropical and subtropical humid forests and deserts. Although charismatic mammals and birds have received more attention because of declining populations and shrinking habitats, reptile species throughout the world are also declining in the Age of Man, the Anthropocene.

As with other taxa, habitat loss is still the greatest threat to most species, especially coupled with fragmentation that isolates populations that manage to survive the development, agriculture, and transportation corridors that pattern human landscapes. Significant other threats come from emerging infectious diseases, for example, ranavirus (affecting land turtles), fibropapillomatosis (affecting sea turtles), and snake fungal disease (particularly in eastern North America). Collection for food or leather, vandalism and malicious killing, and the lethal and sublethal effects of toxic and endocrine-disrupting chemicals also contribute to declines. Because the sex ratio of many reptiles, particularly turtles and crocodilians, is determined by the nest temperature during egg development, there has been growing concern about the effects of increasing temperatures associated with climate change on reproduction and distribution patterns. A recent assessment of the status of the world’s reptiles concluded that 20% were in danger of extinction, and that another 20% were so “data deficient” that we simply do not know how to assess their prospects for survival.

The life histories and conservation status of most of these species are imperfectly understood or completely unknown except for a few of the more charismatic or popular larger species. A surprising amount of what we do know about reptiles is based on a relatively few species, and even then long-term studies are available for only a minute percentage of known species. This is a particularly vexing problem when dealing with species that may not reach sexual maturity for decades (for example, some sea turtles and the tuatara); as might be anticipated, species with long lives and delayed maturity are among the most imperiled species. Even for common species, good population data over long time periods is usually lacking for more than a few locations, and we know little of how populations vary in numbers (what is “natural” variation?), the influences of subtle changes in a species’ community, or the effects of stochastic or periodic disturbances (for example, drought, floods, fire, storms) on population recovery and persistence.

Crocodile by Eelffica. Public domain via Pixabay.

Crocodile alligator dangerous by Eelffica. Public domain via Pixabay.

Another problem is our own human short attention span. We tend not to think in evolutionary time (1000s of years) and we tend to follow the “shifting baseline syndrome,” whereby each human generation becomes accustomed to a slightly more impoverished natural biodiversity. Thus, there is an urgent need for field research on reptiles and their community interactions, and to recognize our own inherent biases in perception and research.

Today, scientists do not just catch snakes, lizards, and turtles to do their research. They employ rather detailed protocols for setting research questions and objectives, and now have an array of sophisticated methods to target critical sampling areas, to identify, collect, and track animals, and to analyze and archive the resulting data. Foremost concerns are to collect data as humanely as possible and to try to minimize sampling biases that have been inherent in many studies, where opportunism seemed to play an all-important role. For most studies, animals are no longer sacrificed to provide data on diet (for example, by using stomach flushing or stable isotopes), reproduction (radiographs, ultrasound), or tissue analysis (blood collection, biopsies for small amounts of tissue). Even detecting the presence of some species might be facilitated by sampling environmental DNA (eDNA), thus minimizing financial expense, difficult logistics, and habitat disturbance when looking for rare or secretive species.

Simply put, reptile ecologists want animals alive in nature in order to follow populations through time. To do this, animals need to be marked so that they can be identified in the future, whether in hand or at some distance. Some animals can be permanently marked by shell notching or scale clipping without causing undue stress (which can be assessed by sampling blood for stress hormones), tagged with Passive Integrated Transponders (PIT tags), or even identified using unique scale or color patterns compared to a digitally assembled computer database. Specialized paints using coded numbers or spots, and even fluorescent powder, can be used to see reptiles (especially lizards and turtles) at a distance without repeatedly disturbing them. Movements are tracked using ever-smaller radio, acoustic, and even satellite transmitters. One of the more exciting recent acoustic discoveries is that hatchling turtles in the Amazon actually use sound to follow adults to feeding grounds.

Analyzing capture data from mark-recapture studies also has become much more rigorous. Reptile ecologists and conservation biologists have moved beyond the notion of simple probability statistics, although they are still useful if understood correctly, to information theoretic approaches that allow more interpretive flexibility to generate testable hypotheses. Reptile scientists now incorporate detection probabilities into assessments of abundance, allowing them a more detailed prediction of abundance and trends in life history parameters, such as population growth rates, recruitment, and survivorship. Multi-state and multi-scale occupancy models allow assessments of status and trends over substantial areas in addition to more localized patterns of extinction and recolonization. Such models let researchers examine factors affecting the presence or absence of a species in a region while accounting for imperfect detection, an important consideration when sampling. GIS and other landscape-scale modelling tools further allow researchers to target areas for surveys and protection. In essence, rapidly-evolving mathematical approaches give researchers tools for assessing the past and predicting the effects of current activities, and for managing for change in the future.

The techniques scientists use to study reptile biology, determine species status and trends, and plans for conservation have moved beyond the simple act of catching an animal. They involve the entire realm of modern science technology, from molecular biology to global modeling. And it all begins with the question of sampling, since unbiased sampling forms the basis for how scientists understand and interpret their data. We appreciate sampling as an integral part of our science, rather than just a means of capturing animals; it is the how and the why and the ‘what does it mean’ foundation of twenty-first-century natural history research.

C. Kenneth Dodd, Jr. is currently Courtesy Associate Professor in the Department of Wildlife Ecology and Conservation at the University of Florida. He has published over 210 research and popular articles, book reviews and book chapters and edited the Amphibian volume in OUP's Techniques in Ecology and Conservation Series. His professional interests are conservation biology, population ecology and demography, monitoring vertebrate populations, sampling approaches, and history of herpetology. Ken lives in Gainesville, Florida, with his wife Marian Griffey and their 8 cats and numerous turtles. He is the editor of Reptile Ecology and Conservation: A Handbook of Techniques (OUP, 2016).