Spatial sampling heterogeneity limits the detectability of deep time latitudinal biodiversity gradients

Area of Science:

• Paleontology
• Biodiversity science
• Geological sciences

Background:

• The latitudinal biodiversity gradient (LBG) describes decreasing species richness from tropics to poles.
• Fossil records suggest LBGs have changed over geological time, but biases hinder recognition.
• Spatial sampling heterogeneity is a major challenge in reconstructing deep-time LBGs.

Purpose of the Study:

• To test the detectability of different latitudinal biodiversity gradient (LBG) types over the last 300 million years.
• To assess the impact of spatial sampling heterogeneity on reconstructing past LBGs.
• To evaluate the effectiveness of sampling standardization in recovering true biodiversity gradients.

Main Methods:

• Utilized a simulation framework to model three LBG types: flat, unimodal, and bimodal.
• Applied spatial sampling windows representative of fossil collection data to simulated LBGs.
• Employed sampling-standardization techniques to analyze reconstructed biodiversity gradients.

Main Results:

• Spatial sampling heterogeneity significantly obscures the detectability of genuine LBGs.
• Known biodiversity patterns were frequently obscured by sampling biases.
• Sampling-standardization improved relative gradient reconstruction but could not eliminate bias from geological/anthropogenic factors.

Conclusions:

• Incomplete and heterogeneous fossil sampling may have led previous studies to misidentify past LBG types, especially between 200 and 20 million years ago.
• These sampling biases can distort global biodiversity estimates and obscure extinction/radiation events.
• Acknowledging and addressing sampling biases is crucial for accurate paleobiogeographic reconstructions.