Clustered versus catastrophic global vertebrate declines

Area of Science:

• Ecology
• Conservation Biology
• Zoology

Background:

• Global vertebrate population declines have been reported, with previous analyses indicating a significant decrease since 1970.
• Existing global mean indices may obscure variations and be sensitive to analytical decisions, potentially misrepresenting overall trends.

Purpose of the Study:

• To investigate the impact of extreme population declines on global vertebrate trend analyses.
• To propose and evaluate an alternative approach for identifying and interpreting population trends, focusing on statistically significant clusters.

Main Methods:

• Re-analyzed existing vertebrate population data, excluding extremely declining populations.
• Developed a novel approach to identify clusters of extreme decline or increase that deviate statistically from the majority of trends.
• Examined trends within specific taxonomic-geographic systems and by animal size and type.

Main Results:

• The previously reported global vertebrate decline of over 50% was driven by less than 3% of populations.
• Excluding these extreme outliers shifted the global trend to an increase.
• Identified 16 systems with extreme decline clusters (around 1% of populations, disproportionately larger animals) and 7 with extreme increase clusters (around 0.4% of populations).
• The majority (98.6%) of populations showed no significant global trend, though specific regions and groups (Indo-Pacific, reptiles, amphibians) showed strong declines with varying certainty.

Conclusions:

• Global mean trends in vertebrate populations are highly sensitive to outliers, necessitating more nuanced analytical approaches.
• Identifying clusters of extreme trends is crucial for accurately interpreting population dynamics and prioritizing conservation efforts.
• Conservation strategies should account for these identified clusters to effectively address regional and taxonomic-specific challenges.