Youtube Video

PM2.5 Air Quality Analyzer

For our CSC 101 final project, we built a data-reporting tool that turns raw PM2.5 air quality readings into a clear, readable summary. The goal was to take CSV data and automatically produce meaningful insights —like averages, extremes, and health-standard exceedances — so it’s easier to understand air quality across different cities over time.

Dataset: PM2.5 readings (µg/m³) from September 2025 – January 2026 across:

Los Angeles
Mexicali
Sacramento

What the program does

At a high level, the program:

Loads one or more city CSV files
Parses each row into a structured data record
Builds a City object for each dataset
Computes key statistics automatically
Writes everything into a generated .txt report

Key outputs include:

Daily / weekly / monthly averages
Highest and lowest PM2.5 readings (with timestamps)
Days exceeding a health standard threshold
A combined summary when multiple cities are included

Sample Program Output:

How to run it

The program is run through command-line arguments (or an IDE run configuration) and includes basic validation so it won’t run unless the inputs make sense.

In our demo:

Running with no arguments prints an invalid arguments message
Running with only a date still fails (because it requires at least one CSV)
Running with a date + one CSV generates a report for that city
Running with a date + multiple CSVs produces a combined report comparing the cities

Program design (Object-Oriented Approach)

We structured the project around an object-oriented design to keep the code readable and scalable.

Core components:

Reading: represents one PM2.5 measurement and its timestamp
City: stores a city’s dataset and computed statistics

When a City object is created, it loads the city’s CSV data and performs the calculations so the object is immediately ready to generate a formatted summary.

What we compute

For each city, the report includes:

Monthly average
Weekly average
Daily average
Highest recorded value + timestamp
Lowest recorded value + timestamp
Days exceeding health standards

Health standard used:
WHO Standards: 5.0 µg/m³

EPA Standards: 9.0 µg/m³

EPA 24hr Standards: 35.0 µg/m³

Testing & reliability

To confirm the analysis stayed correct as the program evolved, we used unit tests with test-case CSV files. This helped verify:

CSV parsing is correct
Average calculations and min/max detection are correct
Output formatting stays consistent

Why it matters

PM2.5 is linked to respiratory health risks, but raw measurement data can be hard to interpret. This project converts many individual readings into an understandable summary so people can more easily compare air quality across major cities and time ranges.

What I learned

This project reinforced:

Designing a clean class model around real datasets
Building a pipeline from raw input → structured objects → analysis → report output
Defensive programming (argument validation + error handling)
Using unit tests to keep logic stable while iterating

Future improvements

If we expanded this project, I’d like to add:

Visualizations (trend lines, comparisons, distributions)
Configurable health standards (switch between WHO/EPA thresholds)
More cities and automated dataset ingestion
Export formats beyond .txt (CSV summary, PDF, or HTML report)

PM 2.5 - CSC 101 Final Project