Classifying Sutherland Nights into Observing-Condition Regimes Using Unsupervised Learning

Project Description:

This project will explore whether nights at the Sutherland observing site can be grouped into a small number of distinct weather and observing-condition regimes using unsupervised machine-learning techniques. Working with archived site data such as temperature, humidity, dew point, wind, and pressure, the student will derive nightly summary features, apply clustering methods, and investigate whether the resulting groups are physically meaningful and potentially related to observing quality. The project offers an introduction to exploratory data analysis, unsupervised learning, and the scientific interpretation of patterns in real astronomical site-monitoring data.

Research Area:

Astronomy

Project Level:

Honours

This Project Is Offered At The Following Node(s):

(UCT)

Special Requirements:

Python programming skills required.