Overview

This research, published in Ecography (2024), introduces a novel modeling approach called BISHOP (Bioclimate Inference from Species' High-resolution Occurrence Patterns) that leverages species occurrence data to predict fine-scale climate variation. The approach addresses a critical gap in ecological research by providing access to high-resolution “bioclimate” data representing microclimatic conditions as experienced by organisms.

Methodological Innovation

BISHOP works by:

• Modeling microclimate as a latent variable in a multi-level Bayesian model that pairs a climate-downscaling function with a niche function predicting species occurrences
• Inferring how physiographic variables (terrain, soil, vegetation) affect local climate conditions, and estimating how these effects vary geographically
• Producing high-resolution (10m) maps of bioclimate over large regions

Empirical Findings

Using data from 216 North American tree species, we:

• Documented biogeographic patterns that enable BISHOP inference
• Estimated the influence of four terrain variables (northness, eastness, windward exposure, and elevational position) on three climate variables
• Generated validated maps of tree-specific bioclimate
• Demonstrated that nearby bioclimates can differ by 5°C in temperature and twofold in moisture
• Revealed that topographically similar landscapes can exhibit dramatically different bioclimate variation

Resources

To accompany this publication, we developed a companion R package that enables users to generate fine-scale tree-focused bioclimate predictions for any landscape in the United States based on the regression models from the paper.

Citation: Kling, M.M., Baer, K.C., & Ackerly, D.D. (2024). A tree’s view of the terrain: downscaling bioclimate variables to high resolution using a novel multi-level species distribution model. Ecography, e07131. https://doi.org/10.1111/ecog.07131