EPA Air-Quality Exploration
This research project used Python and EPA air-quality data to examine what changes when data collection is extended from monitoring toward explanatory or predictive modeling.
Translating the distinction between monitoring and modeling into a broader systems-research question.
Foundation
The problem
Monitoring can describe observed conditions, while modeling attempts to explain, estimate, or predict behavior. These activities require different assumptions and support different claims.
Why it matters
Confusing observation with explanation can lead researchers to make claims that exceed the evidence available in the data.
Current system
How it currently works
The present model
EPA air-quality measurements provide the observational data.
Python is used for cleaning, organizing, and visualizing data.
Monitoring and modeling are treated as different epistemic activities.
Assumptions are examined before explanatory claims are made.
Inquiry
Open investigation
Questions guiding the work
What can monitoring establish without a model?
What assumptions does modeling introduce?
How should missing or uneven measurements be interpreted?
Epistemic state
Current understanding
What appears known—and what does not
What I presently understand
A dataset does not explain itself.
Monitoring and modeling support different kinds of conclusions.
Visualization can reveal patterns without establishing causality.
What remains unresolved
How much environmental context is required for meaningful modeling?
Which variables are necessary for causal interpretation?
Progress
Development history
Project timeline
Research reflection
Documented the research process and methodological lessons.
Research presentation
Presented the distinction between Monitor Only and Monitor and Model.
Research program began
Started the ASRI Python and environmental data research track.
Connections
Related work