Getting the most out of ecotoxicological field data with generalized linear and additive mixed models: a six-step...

Area of Science:

• Ecotoxicology
• Ecology
• Environmental Risk Assessment
• Statistical Modeling

Background:

• Analyzing field ecotoxicological data is challenging due to natural system variability and complex interactions.
• Generalized linear and additive mixed models (GLMM, GAMM) are powerful tools for ecological data but are often perceived as complex.
• Hesitation exists in using GLMM and GAMM results for regulatory chemical risk assessment.

Purpose of the Study:

• To present a framework that demystifies the development and application of GLMM and GAMM for ecotoxicological field study data.
• To enhance the understanding and adoption of these advanced statistical models in regulatory evaluations.
• To demonstrate the utility of GLMM and GAMM in extracting maximum insights from noisy ecotoxicological datasets.

Main Methods:

• A six-step framework is proposed, including data exploration, model calibration, internal validation, selection, Minimum Detectable Difference (MDD) evaluation, and interpretation.
• The framework allows for checking the significance of treatment-related effects at two independent stages.
• The methodology is exemplified using a case study involving common voles exposed to a fungicide under field conditions.

Main Results:

• The case study demonstrated the flexibility of GLMM and GAMM in handling diverse data types (counts, proportions, continuous) and incorporating repeated sampling.
• These models effectively addressed potential non-linear dynamics and multiple influencing factors in ecotoxicological endpoints.
• The framework facilitated a more robust analysis of ecotoxicological field data, overcoming challenges of natural system variability.

Conclusions:

• GLMM and GAMM are highly advantageous for analyzing complex ecotoxicological field data, offering flexibility and robust statistical power.
• The presented framework simplifies the application of these models, promoting their use in regulatory risk assessment.
• This approach enables more reliable interpretation of chemical risks by effectively disentangling multiple environmental factors.