Researchers as predators: are field biologists altering traits of the populations that they study?

Field biologists rarely directly measure the long-term impact of their work on the populations they study. However, heavy or repeated sampling (with or without release) could potentially alter population traits and even bias study conclusions. Trinidadian guppies (Poecilia reticulata) are a frequently-studied species in evolutionary biology, particularly in the context of variation in predation. Here, we tested how research sampling intensity might be associated with behavioral traits (boldness, exploration, and shoaling), coloration (relative black and orange area), or size (body mass) in two low predation guppy populations. We carried out a literature review and identified paired sites in the Aripo and Paria rivers that have been either weakly-sampled or heavily-sampled for research, collected guppies from these sites, and measured traits. None of the behavioral traits, orange colouration, or mass showed any detectable impact of sampling intensity, which is reassuring for long-term guppy research. However, guppies from heavily-sampled sites had greater relative black area than did those from weakly-sampled sites. This last result suggests that guppy researchers could have unintended impacts on some guppy traits, and highlights the importance of considering long-term researcher-induced effects in study design.