Statistics matter: data aggregation improves identification of community-level effects compared to a commonly used...

Area of Science:

• Environmental Toxicology
• Ecotoxicology
• Community Ecology

Background:

• Identifying toxicant effects on complex biological communities is challenging due to natural variability.
• The trait-based SPEAR (Species At Risk) approach was developed to identify vulnerable taxa and aggregate them, reducing noise and improving detection.
• Further quantitative investigation is needed for the SPEAR approach's application in mesocosm studies and ecological risk assessment.

Purpose of the Study:

• To quantitatively evaluate the potential of taxa aggregation within the SPEAR approach for mesocosm data analysis.
• To compare the effectiveness of taxa aggregation against traditional multivariate methods like redundancy analysis.
• To assess the utility of the SPEAR approach in ecological risk assessment of toxicants.

Main Methods:

• Analysis of empirical data from a long-term mesocosm experiment using stream invertebrates and an insecticide.
• Utilisation of simulated datasets with varying data matrix saturation, replication numbers, and between-replicate differences.
• Comparison of the taxa aggregation method with the redundancy analysis (RDA) multivariate statistical method.

Main Results:

• The taxa aggregation approach detected toxicant effects at lower data matrix saturation and with fewer replicates compared to redundancy analysis.
• This method proved more robust in the presence of higher between-replicate differences.
• The SPEAR approach demonstrated up to 1,000 times greater sensitivity in detecting long-term effects at lower toxicant concentrations.

Conclusions:

• Taxa aggregation methods, like the SPEAR approach, significantly enhance the sensitivity and reliability of toxicant effect detection in ecological studies.
• These methods are particularly valuable for mesocosm studies, which often suffer from poor replication and high variability.
• The SPEAR approach shows strong potential for improving ecological risk assessment by enabling detection of subtle, long-term toxicant impacts.