From gaps to insights: quantifying biological data uncertainty and revisiting biodiversity in environmental space

Abstract

Macroecology seeks to understand the ecological and evolutionary processes that influence biodiversity patterns across broad spatial and temporal scales. However, this goal is limited by substantial gaps in information on species’ geographic distributions, environmental niches, and phylogenetic relationships, as well as by the challenge of quantifying the uncertainty associated with these data. Moreover, many macroecological hypotheses receive inconsistent empirical support, partly because they are commonly tested in geographic space, where historical contingencies, biogeographic processes, and biotic interactions may obscure the effects of climate-driven ecological processes. This thesis presents methodological and conceptual advances to address these two major challenges in macroecology. First, I develop approaches to quantify uncertainty across multiple dimensions of biodiversity data, including species’ geographic distributions, climatic niches, and phylogenetic histories, together with computational tools implemented in R to facilitate their broader application by the scientific community. Second, I investigate macroecological hypotheses directly in environmental space, an approach that minimizes the influence of geographic historical contingencies by treating equivalent climatic conditions as independent sampling units. Within this framework, I evaluate predictions of Bergmann’s rule for body-size patterns and propose a new interpretation of the mid-domain hypothesis based on the geometry of species’ geographic distributions and environmental niches. Together, these contributions advance our understanding of how limitations in biodiversity data affect macroecological inference and how environmental space can provide new insights into the ecological mechanisms influencing biodiversity patterns.