Population variability complicates the accurate detection of climate change responses

Abstract

The rush to assess species' responses to anthropogenic climate change (CC) has underestimated the importance of interannual population variability (PV). Researchers assume sampling rigor alone will lead to an accurate detection of response regardless of the underlying population fluctuations of the species under consideration. Using population simulations across a realistic, empirically based gradient in PV, we show that moderate to high PV can lead to opposite and biased conclusions about CC responses. Between pre- and post-CC sampling bouts of modeled populations as in resurvey studies, there is: (i) A 50% probability of erroneously detecting the opposite trend in population abundance change and nearly zero probability of detecting no change. (ii) Across multiple years of sampling, it is nearly impossible to accurately detect any directional shift in population sizes with even moderate PV. (iii) There is up to 50% probability of detecting a population extirpation when the species is present, but in very low natural abundances. (iv) Under scenarios of moderate to high PV across a species' range or at the range edges, there is a bias toward erroneous detection of range shifts or contractions. Essentially, the frequency and magnitude of population peaks and troughs greatly impact the accuracy of our CC response measurements. Species with moderate to high PV (many small vertebrates, invertebrates, and annual plants) may be inaccurate 'canaries in the coal mine' for CC without pertinent demographic analyses and additional repeat sampling. Variation in PV may explain some idiosyncrasies in CC responses detected so far and urgently needs more careful consideration in design and analysis of CC responses.

Seasonal shifts of biodiversity patterns and species’ elevation ranges of butterflies and moths along a complete rainforest elevational gradient on Mount Cameroon

“…Temporal changes of diversity along tropical elevational gradients have become a widely used approach for studying the global change impacts on biota, despite the above‐mentioned insufficient knowledge of their seasonal dynamics (McCain, Szewczyk, & Knight, ). The current consensus predicts that the ongoing global climate change will cause uphill shifts of tropical montane species, due to their relatively narrow thermal tolerances (Forero‐Medina et al, ; Laurance et al, ), resulting in subsequent extinctions of mountaintop species (Colwell et al, ).…”

Moving beyond presence and absence when examining changes in species distributions

“…Regardless of whether climate change impacts are assessed using abundance or presence, it is difficult to be confident of range shifts if populations are variable (McCain et al, 2016). Studies that are based on too few points in time are likely to overestimate change when they confound long-term trends with short-term variability (Brown et al, 2016).…”

Characterizing opportunistic breeding at a continental scale using all available sources of phenological data: An assessment of 337 species across the Australian continent

“…Further, the typical practice of looking for changes in the start and conclusion of breeding in response to climate change may not be appropriate in areas with extensive breeding periods. Assessments of species responses to climate change need to include measures of population variation (McCain et al 2016); this is especially important for areas of the world that experience a high level of inter-annual climatic variation.…”