Air Quality Data Analysis Project with Python

"""Clean and summarise synthetic air-sensor measurements."""

from __future__ import annotations

import numpy as np
import pandas as pd


def make_demo_air_quality(days: int = 365, seed: int = 42) -> pd.DataFrame:
    rng = np.random.default_rng(seed)
    dates = pd.date_range("2025-01-01", periods=days)
    x = np.arange(days)
    pm25 = 34 + 11 * np.sin(2 * np.pi * x / 90) + rng.normal(0, 6, days)
    no2 = 26 + 5 * np.sin(2 * np.pi * x / 7) + rng.normal(0, 3, days)
    data = pd.DataFrame({"date": dates, "pm25": np.clip(pm25, 1, None), "no2": np.clip(no2, 1, None)})
    data.loc[rng.choice(days, 18, replace=False), "pm25"] = np.nan
    return data


def analyse_air_quality(data: pd.DataFrame, demo_pm25_threshold: float = 40.0) -> tuple[pd.DataFrame, dict[str, float]]:
    air = data.copy().sort_values("date")
    air["pm25"] = air["pm25"].interpolate(limit_direction="both")
    air["month"] = pd.to_datetime(air["date"]).dt.to_period("M").astype(str)
    monthly = air.groupby("month", as_index=False).agg(mean_pm25=("pm25", "mean"), p90_pm25=("pm25", lambda s: s.quantile(0.9)), mean_no2=("no2", "mean"))
    kpis = {"mean_pm25": float(air["pm25"].mean()), "days_above_demo_threshold": float((air["pm25"] > demo_pm25_threshold).sum()), "rows": float(len(air))}
    return monthly, kpis


def main() -> None:
    monthly, kpis = analyse_air_quality(make_demo_air_quality())
    print({k: round(v, 2) for k, v in kpis.items()})
    print(monthly.round(2).to_string(index=False))
    print("Synthetic sensor data; the 40-unit threshold is illustrative, not health guidance.")


if __name__ == "__main__":
    main()