Tuesday 04 March 2025
A team of astronomers has developed a new way to classify galaxies, one that’s more comprehensive and accurate than traditional methods. The approach uses principal component analysis (PCA) to identify patterns in the visible light spectra of nearby galaxies.
Galaxies are incredibly diverse, ranging from spiral-shaped systems with vibrant star-forming regions to elliptical ones with little to no activity. To better understand these differences, astronomers have long used diagnostic diagrams like the Baldwin-Phillips-Terlevich (BPT) diagram to categorize galaxies based on their emission-line properties.
However, this traditional approach has its limitations. For example, it relies on a limited set of emission lines and can be biased against certain types of activity. The new method, developed by scientists in Spain and the UK, addresses these issues by projecting galaxies onto a 2D latent space defined by the first three principal components (PCs) of their entire visible spectra.
The team applied this approach to a sample of over 68,000 nearby lenticular (S0) galaxies from the Sloan Digital Sky Survey. By analyzing the resulting diagram, known as ΔPS-PC3, they found that it effectively captured the full range of galaxy activity, including objects systematically excluded from traditional classifiers.
One key advantage of this new method is its ability to identify galaxies with low levels of activity, which are often missed by traditional approaches. The team discovered that many of these inactive galaxies exhibit spectra resembling those of LINER (Low-Ionization Nuclear Emission-Line Region) galaxies, a type previously thought to be relatively rare.
The study also shed light on the prevalence of different types of galaxy activity. By analyzing the probabilities of class membership for each galaxy, the researchers found that most present-day active S0s fall into the star-forming and composite classes. However, they also detected a significant number of galaxies with LINER-like spectra, which were previously unknown or misclassified.
The implications of this work are far-reaching. By providing a more comprehensive understanding of galaxy activity, astronomers can better understand how galaxies evolve over time and how supermassive black holes influence their development. This new approach could also be applied to other types of galaxies, such as elliptical ones, to gain further insights into the universe.
The development of this PCA-based classification method is an important step forward in understanding galaxy diversity. It highlights the power of innovative statistical techniques in astronomy and demonstrates the value of collaborative research between scientists from different institutions.
Cite this article: “Revolutionizing Galaxy Classification with Principal Component Analysis”, The Science Archive, 2025.
Galaxies, Classification, Pca, Astronomy, Visible Light Spectra, Principal Component Analysis, Galaxy Activity, Liner Galaxies, Sloan Digital Sky Survey, Galaxy Evolution
