The consequences of null model selection for predicting mortality from multiple stressors.

Area of Science:

• Ecology
• Environmental Science
• Toxicology

Background:

• Ecological systems face increasing multiple stressors, necessitating evaluation of their joint impacts.
• Null models, assuming independent stressor action, are used to predict these joint effects in data-limited scenarios.
• Selecting appropriate null models is critical for accurate ecological risk assessment.

Purpose of the Study:

• To evaluate the performance of five null models in predicting mortality from dual stressor exposure.
• To determine if a universally precautionary null model exists for multiple stressor scenarios.
• To identify key factors influencing the accuracy of null model predictions.

Main Methods:

• A simulation study comparing five distinct null models for predicting mortality.
• A meta-analysis re-analyzing existing data using a multi-model framework.
• Statistical analysis to identify predictors of joint stressor effect magnitude.

Main Results:

• Null models exhibited substantial variation in predicted mortality, differing by up to 67.5% higher or 50% lower than the Simple Addition model.
• No single null model consistently predicted the highest mortality across all simulations, negating a universal 'precautionary' model.
• 54% of observed joint effects fell within the range of null model predictions, but most models underestimated effects.
• Individual stressor intensity was the strongest predictor of joint effect magnitude.

Conclusions:

• The choice of null model significantly impacts predictions of multiple stressor effects in ecological systems.
• Existing null models often underestimate joint stressor impacts, highlighting the need for cautious interpretation.
• Characterizing individual stressor effects remains essential for accurate prediction of combined impacts and effective ecosystem management.