Faculty Seed Grant Projects


Scaling climate change vulnerability from individuals to species: how do assessment models compare? 


As climate change outpaces the rate at which climate vulnerability and risk are assessed for most taxa, streamlining vulnerability assessments for multiple taxa is rapidly becoming a scientific imperative (1). Vulnerability assessments are performed along a continuum of broad-scale, coarse-grain approaches to data-intensive, individual-based simulations (Fig. 1). Advances at both ends of the continuum (2, 3) are improving our ability to quantify species’ vulnerability to climate change at specific scales, from general to precise; yet, the variability of outcomes from these assessments is poorly understood. Quantifying how vulnerability assessments scale across approaches will be critical as the pace of climate change requires a variety of tools to evaluate species’ status.

To address this knowledge gap, we propose leveraging a suite of approaches and resources to compare assessments of multiple anuran (frog and toad) species in the United States. Specifically, we ask: how does our understanding of climate change vulnerability vary by geographic scale, model precision, and natural histories? To address this, we propose two aims:

  1. Complete a trait database for anurans of the United States; characterize life history strategies, ecological associations, and intraspecific trait variation that emerge; and identify strategies and species with high intrinsic sensitivity to climate change. 
  2. Identify a subset of well-studied species (N=3) spanning a range of natural strategies for which species distribution models (SDMs), mechanistic niche models (MDMs), and spatially explicit individual-based models (SIBMs) will be developed to test the degree of convergence of vulnerability assessments (Figure 1). 

Figure 1. A) Trade-offs between types of approaches to assess vulnerability of species to climate change. Multispecies, range-wide approaches, such as traits-based analyses and SDMs, are applicable to a wide range of species but are often low precision; in contrast, data-intensive models, such as SIBMs, offer high precision but are not applicable for many data-poor species. B) Comparing outcomes for well-studied species (N=3) will help evaluate the degree to which vulnerability outcomes may converge (“High agreement” scenario) or diverge (“Low agreement”) across assessment approaches.

References: (1) Williams et al. 2008 PloS Biology 6:2621-2626, (2) Pacifici et al. 2017 Nature Climate Change 7:205-208, (3) Jager et al. 2018 Ecological Modelling 384:341-352, (4) Murray et al. 2002 Austral Ecology 27:291-310.