Mapping diversity through geographical, environmental and attribute spaces

Understanding biodiversity patterns requires integrating information across the interrelated dimensions that organize life. Species occupy specific regions in geographic space, respond to environmental conditions and vary within attribute (or trait) spaces that are shaped by their ecological and evolutionary histories. However, most analyses address these dimensions independently, limiting our ability to understand how constraints in one space interrelate to patterns in another. Developing approaches that explicitly quantify and connect the geometry of these spaces should unify the study of biodiversity across scales and dimensions. We implement such a framework in the letsR R package by constructing a triad of presence–absence matrices (PAMs) representing geographic, environmental and attribute spaces. A concise set of functions computes geometry‐aware descriptors such as centrality, frequency, isolation and border proximity, and allows these descriptors to be propagated among spaces through explicit connectors. This design builds on the letsR architecture and integrates with functions to facilitate macroecological analysis. Using global terrestrial mammals as a model group, we illustrate how descriptors derived from environmental space can be mapped back into geographic and attribute spaces. This approach reveals where particular regions of climatic space are common or rare, how isolated they are and how these properties influence species richness across dimensions. This novel implementation lowers the barrier to reproducible, geometry‐aware analyses and enables direct tests of how constraints in one space influence diversity in others, advancing a more integrated understanding of macroecological and macroevolutionary patterns.