Cataloguing species in a lowland Amazon rainforest

by Rudolf von May


A Neotropical water snake (Hydrops), one of several reptiles recently recorded in southeast Peru. Photo courtesy of Emanuele Biggi and Francesco Tomasinelli.

Most researchers who study biodiversity agree that a lot of time and effort are needed to wholly characterize any tropical rainforest community. For herpetologists, it is well understood that it takes years to figure out how many species of reptiles and amphibians live in a tropical rainforest site. Snakes and caecilians are particularly challenging, because many of these species are hard to find due to their cryptic habits. One way to assess how we are progressing in any biotic survey is to look at species accumulation curves, that is, the number of species observed versus the sampling effort (here, effort can be expressed in number of plots surveyed, number of individuals observed, number of trap days, or other equivalent measures). When our species accumulation curve approaches a plateau, we gain a perspective of “what’s there” and, in turn, can reasonably estimate the actual number of species that live in a particular place. In fact, from the shape of accumulation curves we can even make predictions before a curve begins to plateau. However, this does not identify what the species will be.


Species accumulation curves based on surveys of lowland rainforest frogs in Peru. Two methods were used, visual encounter surveys (Transects) and leaf-litter plots (Plots), and each curve represents the number of species for a given number of observed individuals (data from von May et al. 2009). The bars indicate  standard deviation after rarefaction. In general, a species accumulation curve for snakes would be similar, though the rate of new species added to the list would be lower and might vary depending on the method used.

Thus, what do you do if you have been surveying a tropical rainforest site for several years and you think your species accumulation curve has approached a plateau? The first question you may want to ask is whether the curve has really reached a plateau. If we assume that our field methods are effective, we might still be missing several species for other reasons. For instance, some species might exclusively live in the canopy, or some may have unusual activity patterns. Should we invest more time surveying the site? Perhaps bring more herpetologists? Try new survey methods?

All of the above recently happened at the Los Amigos Biological Station in southeast Peru, where 12 herpetologists, including seven members of the Rabosky Lab (University of Michigan Museum of Zoology and Department of Ecology and Evolutionary Biology), spent several weeks conducting fieldwork earlier this year. Herpetological studies at Los Amigos started over a decade ago, and, by the time we traveled to the region, expectations regarding how many species we would find were diverse. Our general goal was to collect different types of data that could be used to better understand why so many species coexist in lowland Amazonian forests.


Team members Iris Holmes, Mike Grundler, Pascal Title, and Mark Cowan, getting ready for some rainforest action. The tiger and the lion guarding the sidewalk belong to a local business in Puerto Maldonado, the capital of Madre de Dios region in southeast Peru. Although these cats obviously don’t live in the wild in South America, the team encountered jaguars and ocelots in the field. Photo by R. von May.

While we were preparing for the trip in Michigan, several colleagues working in Amazonian Peru reported that it was unusually warm for January. As you may imagine, it is a bit hard to grasp such news in winter while watching a snow plow clearing the streets outside. But then you realize that this, according to NOAA and similar organizations, has been one of the “top” El Niño Southern Oscillation events on modern record. This increase in temperature was also confirmed by measurements taken at our field site. Thanks to the Amazon Conservation Association (ACA), daily temperature and rainfall have been measured, and records kept, since the year 2000. As we can see in the figure included below, maximum air temperatures were consistently high over the first half of 2016. This spike in temperature started around mid-November 2015 and continued till mid-May 2016. Over this period, maximum air temperatures of 30 °C (= 96 °F) or higher were very common (70% of daily records), and 13 days had maximum air temperatures above 38 °C (= 100 °F). It’s likely that for some people, these temperatures are not a big deal compared with what they have experienced in other regions. However, high temperatures are detrimental for many rainforest organisms.


Daily air temperature measured at Los Amigos Biological Station, Peru. Red color indicates maximum air temperature. The dashed line placed at 30 °C is used to highlight the temperature increase in 2016. Though no data were available for two periods (gaps in chart), there is a marked pattern of seasonal fluctuations: the warmest period on most years is between October and March. Figure by Rudolf von May; data courtesy of Amazon Conservation Association.

How has this increase in temperature affected wildlife? And what has been the impact of recent droughts on Amazonian forests? Recent reports focusing on tree communities have shown that plant growth throughout the Amazon slowed down during recent drought periods, whilst overall tree mortality rates increased. This in turn affects the overall forest productivity and may impact the long-term ability of Amazon forests to function as a carbon sink. However, other broad-spectrum effects of increased temperatures (or more frequent droughts) on Amazonian biodiversity, which contains about 12,000 tree species alone, remain far from being understood.

Long-term research at selected sites will shed light on how the reptile and amphibian communities cope with these environmental changes. Los Amigos Biological Station and the larger Los Amigos Conservation Concession enable access to major Amazonian habitats such as terra firme and floodplain forests, which are interlaced by rivers, oxbow lakes, streams, and pools that support millions of reptiles and amphibians. Because Los Amigos Conservation Concession borders Manu National Park to the west, the preserve also protects key lowland habitats of a broader elevational gradient that is home to the most diverse herpetological communities on Earth. Furthermore, the preserve protects a large area of old-growth Amazon rainforest that, together with Manu, is critically important for several indigenous groups living in the region.


The ringed caecilian, Siphonops annulatus, is one of several amphibians that are rarely seen in the field in western Amazonia. With a fossorial lifestyle, they spend most of their time underground and exhibit an unusual type of parental care in which females produce a specialized skin that serves as food for their offspring. Photo by R. von May.

Though we didn’t find any caecilians during this trip, our field surveys added nine snake species to the checklist of Los Amigos. Most of these new records resulted from the use of a method that was not used in previous years—hence producing a change in trajectory, or a “bump,” in the overall species accumulation curve for reptiles. Furthermore, while back in Michigan, I kept in touch with two colleagues, Emanuele Biggi and Francesco Tomasinelli, who were conducting fieldwork at Los Amigos in July 2016. They were delighted to report a great find, a Neotropical water snake in the genus Hydrops (see top photo). Although the identification of this species is pending, it represents a new record for the station. Considering this year’s results, it is reasonable to expect that future surveys will uncover additional species as well as enable us to better understand the impacts of environmental variability in this truly remarkable environment.

Museum data sheds light on coral snakes and their mimics

A U-M led team of herpetologists has just published a pair of papers dealing with the evolution of warning coloration and mimicry in New World snakes. The first is published in Nature Communications and deals with the evolutionary dynamics of coral snake-like coloration (red-and-black banding) in both space and time. The University of Michigan released a really nice multimedia feature on this paper, which includes some great video footage from our recent UMMZ Herpetology expedition to the Peruvian Amazon. One of my favorite subclips from the accompanying video comes at 1:55, where you can see some footage we shot of a harmless litter-dwelling snake (Atractus elaps) doing a spectacular imitation of a coral snake defensive display (see below). This display was quite convincing and inspired a considerable level of caution among our group of professional herpetologists!

Continue reading