Python Weather Data Scraping, CSV Export, and Visualization with Matplotlib

The article explains a complete workflow for obtaining weather information from the China Weather website using Python. It first requests the 7‑day and 15‑day pages, parses the HTML with BeautifulSoup , and extracts timestamps, temperature, humidity, wind direction, wind level, precipitation, and air‑quality index.

Two main functions are provided: getHTMLtext(url) fetches the raw page content, while get_content(html) processes the 7‑day data and get_content2(html) handles the 15‑day forecast. The extracted values are stored in Python lists and then written to CSV files ( weather1.csv for the current day and weather14.csv for the next 14 days) using the csv module.

After data collection, the script loads the CSV files with pandas.read_csv and creates several visualizations:

A 24‑hour temperature curve with maximum, minimum, and average lines.

A 24‑hour relative humidity curve with similar annotations.

An air‑quality (AQI) bar chart that colors each hour according to the AQI level.

A polar radar chart showing wind direction distribution and average wind speed.

A scatter plot of temperature versus humidity with a calculated correlation coefficient.

A 14‑day high/low temperature line chart and a pie chart of weather condition frequencies.

All visualizations are built with matplotlib.pyplot , and helper functions such as calc_corr compute statistical measures. The code also includes utilities for converting Chinese wind direction strings to degrees for the radar plot.

The article concludes with observations: temperature and humidity are strongly negatively correlated; air quality is influenced by temperature inversions and diurnal cycles; wind patterns show no dominant direction over the 14‑day period; and the 14‑day forecast indicates a cooling trend after day 8.

Finally, the full source code is provided, organized into three modules: weather.py (data crawling and CSV export), data1_analysis.py (visualization of the current day's data), and data14_analysis.py (visualization of the 14‑day forecast). The code snippets are reproduced verbatim within ... blocks to preserve formatting.