meteorology-driver-classification

This skill provides a structured framework for categorizing complex environmental and meteorological datasets into meaningful driver groups. Designed for environmental scientists, hydrologists, and data analysts, the skill facilitates rigorous attribution analysis by ensuring that raw inputs—such as air temperature, precipitation, wind patterns, and anthropogenic factors—are systematically organized. By standardizing the classification of variables into core categories like Heat, Flow, Wind, and Human activities, the agent can perform more accurate statistical grouping and causal inference within environmental systems.

• Categorize variables into standard physical domains: Heat (thermal/radiation), Flow (hydro-logical movement), Wind (circulation/pressure), and Human (anthropogenic impacts).

• Implement custom derived variables, such as calculating Net Radiation from longwave and shortwave components.

• Ensure data integrity by validating that variables within the same category are appropriately correlated and physically consistent.

• Support complex attribution workflows by identifying which environmental drivers most significantly impact observed changes in ecosystems or climate patterns.

• Input Requirements: Expects tabular datasets (e.g., DataFrames) containing raw measurements such as temperature, humidity, cloud cover, inflow/outflow, and land use metrics.

• Output Expectations: Provides a mapped structure where each raw variable is assigned a driver label, often resulting in consolidated features suitable for statistical models.

• Best Practices: Always verify that groupings are mutually exclusive to prevent collinearity issues during downstream analysis.

• Constraint Handling: When variables do not fit cleanly, utilize domain-specific knowledge to define the most relevant physical driver or create a new derived metric. Keep the total number of categories constrained between 3 and 5 to maintain model interpretability and reduce noise in attribution tasks. Always document the rationale behind specific grouping decisions to ensure reproducibility.