Spatial analysis improves species distribution modelling during range expansion

Area of Science:

• Ecology
• Biogeography
• Computational Biology

Background:

• Species distribution models (SDMs) traditionally assume equilibrium between species ranges and environmental conditions.
• This equilibrium assumption is often unmet due to limited dispersal and spatial autocorrelation in environmental factors.
• Non-equilibrium dynamics are crucial for understanding species' responses to environmental changes, particularly climate change.

Purpose of the Study:

• To evaluate the performance of SDMs when coupled with spatial eigenvector mapping under non-equilibrium conditions.
• To simulate species' range expansion under two distinct non-equilibrium scenarios.
• To determine the importance of spatial variables in SDMs during range expansion.

Main Methods:

• Development and application of a simulation model for species' range expansion.
• Coupling of Species Distribution Models (SDMs) with spatial eigenvector mapping.
• Evaluation of model fit and the relative importance of spatial variables across different scenarios.

Main Results:

• Models incorporating spatial variables demonstrated the highest goodness-of-fit.
• The influence of spatial variables varied significantly between the two simulated non-equilibrium scenarios.
• The inclusion of spatial factors proved particularly important under the colonization-lag non-equilibrium scenario.

Conclusions:

• Spatial modeling effectively captures mechanisms driving range cohesion and species distribution under climate change.
• Integrating spatial components enhances SDM performance, especially when non-equilibrium dynamics are present.
• SDMs need to account for complex, dynamic scenarios to accurately predict species distributions.