A non-parametric cause-effect testing for environmental variables - method and application.

Area of Science:

• Environmental Science
• Ecology
• Statistics

Background:

• Traditional similarity indexes and regression methods struggle to establish causality between environmental variables, especially with limited data.
• Existing statistical approaches often rely on data distribution assumptions, limiting their applicability in real-world environmental analyses.

Purpose of the Study:

• To develop a novel, non-parametric method for quantifying the strength of cause and effect (SCE) between environmental variables.
• To apply this method to empirical European Union environmental data to identify causal relationships.
• To establish critical levels for hypothesis testing of causal links in environmental data.

Main Methods:

• Devised a new statistic, the strength of cause and effect (SCE), to measure causal relationships.
• Utilized empirical environmental data from the European Union for analysis.
• Constructed a ranking space and calculated statistic distributions to define critical values for hypothesis testing.

Main Results:

• Identified direct causal links between certain sectoral activities and environmental quality (water and air).
• Demonstrated a significant and clear cause-effect relationship between water quality and biodiversity.
• The developed SCE statistic proved effective in analyzing environmental data without strict distribution assumptions.

Conclusions:

• The novel SCE method offers a robust approach for analyzing causality in environmental science, particularly with small datasets.
• Policy-relevant insights were generated regarding the direct impacts of human activities on environmental quality and biodiversity.
• Findings underscore the interconnectedness of environmental factors and provide a basis for informed environmental policy-making.