Principal Component Analysis Tomography Research Articles

Principal component analysis tomography in near-infrared integral field spectroscopy of young stellar objects – I. Revisiting the high-mass protostar W33A

ABSTRACT W33A is a well-known example of a high-mass young stellar object showing evidence of a circumstellar disc. We revisited the K-band NIFS/Gemini North observations of the W33A protostar using principal components analysis tomography and additional post-processing routines. Our results indicate the presence of a compact rotating disc based on the kinematics of the CO absorption features. The position–velocity diagram shows that the disc exhibits a rotation curve with velocities that rapidly decrease for radii larger than 0.1 arcsec (∼250 au) from the central source, suggesting a structure about four times more compact than previously reported. We derived a dynamical mass of 10.0$^{+4.1}_{-2.2}$ $\rm {M}_\odot$ for the ‘disc + protostar’ system, about ∼33 per cent smaller than previously reported, but still compatible with high-mass protostar status. A relatively compact H2 wind was identified at the base of the large-scale outflow of W33A, with a mean visual extinction of ∼63 mag. By taking advantage of supplementary near-infrared maps, we identified at least two other point-like objects driving extended structures in the vicinity of W33A, suggesting that multiple active protostars are located within the cloud. The closest object (Source B) was also identified in the NIFS field of view as a faint point-like object at a projected distance of ∼7000 au from W33A, powering extended K-band continuum emission detected in the same field. Another source (Source C) is driving a bipolar $\rm {H}_2$ jet aligned perpendicular to the rotation axis of W33A.

NGC 1566: analysis of the nuclear region from optical and near-infrared Integral Field Unit spectroscopy

We analysed the centre of NGC 1566, which hosts a well-studied active galactic nucleus (AGN), known for its variability. With the aid of techniques such as Principal Component Analysis Tomography, analysis of the emission-line spectra, channel maps, Penalized Pixel Fitting and spectral synthesis applied to the optical and near-infrared data cubes, besides the analysis of Hubble Space Telescope images, we found that: (1) the AGN has a Seyfert 1 emission, with a very strong featureless continuum that we described as a power law with spectral index of 1.7. However, this emission may come not only from the AGN [as its point spread function (PSF) is broader than the PSF of the broad-line region (BLR)], but from hot and young stars, the same ones that probably account for the observed sigma-drop. (2) There is a correlation between redshift and the full width at half-maximum of the BLR emission lines. With a simple model assuming gravitational redshift, we described it as an emitting ring with varying emitting radii and small inclination angles. (3) There is an H II region close to the AGN, which is composed of many substructures forming an apparent spiral with a velocity gradient. (4) We also detected a probable outflow coming from the AGN and it seems to contaminate the H II region emission. (5) We identified an H2 rotating disc with orientation approximately perpendicular to this outflow. This suggests that the rotating disc is an extension of an inner torus/disc structure, which collimates the outflow emission, according to the Unified Model.

Integral field unit spectroscopy of 10 early-type galactic nuclei – I. Principal component analysis Tomography and nuclear activity

Most massive galaxies show emission lines that can be characterized as LINERs. To what extent this emission is related to AGNs or to stellar processes is still an open question. In this paper, we analysed a sample of such galaxies to study the central region in terms of nuclear and circumnuclear emission lines, as well as the stellar component properties. For this reason, we selected 10 massive ($\sigma$ > 200 km/s) nearby (d < 31 Mpc) galaxies and observed them with the IFU/GMOS (integral field unit/Gemini Multi-Object Spectrograph) spectrograph on the Gemini South Telescope. The data were analysed with principal component analysis (PCA) Tomography to assess the main properties of the objects. Two spectral regions were analysed: a yellow region (5100-5800 A), adequate to show the properties of the stellar component, and a red region (6250-6800 A), adequate to analyse the gaseous component. We found that all objects previously known to present emission lines have a central AGN-type emitting source. They also show gaseous and stellar kinematics typical of discs. Such discs may be co-aligned (NGC 1380 and ESO 208 G-21), in counter-rotation (IC 1459 and NGC 7097) or misaligned (IC 5181 and NGC 4546). We also found one object with a gaseous disc but no stellar disc (NGC 2663), one with a stellar disc but no gaseous disc (NGC 1404), one with neither stellar nor gaseous disc (NGC 1399) and one with probably ionization cones (NGC 3136). PCA Tomography is an efficient method for detecting both the central AGN and gaseous and stellar discs. In the two cases (NGC 1399 and NGC 1404) in which no lines were previously reported, we found no evidence of either nuclear or circumnuclear emission, using PCA Tomography only.

IFU spectroscopy of 10 early-type galactic nuclei – II. Nuclear emission line properties

Although it is well known that massive galaxies have central black holes, most of them accreting at low Eddington ratios, many important questions still remain open. Among them, are the nature of the ionizing source, the characteristics and frequencies of the broad line region and of the dusty torus. We report observations of 10 early-type galactic nuclei, observed with the IFU/GMOS spectrograph on the Gemini South telescope, analysed with standard techniques for spectral treatment and compared with results obtained with principal component analysis Tomography (Paper I). We performed spectral synthesis of each spaxel of the data cubes and subtracted the stellar component from the original cube, leaving a data cube with emission lines only. The emission lines were decomposed in multi-Gaussian components. We show here that, for eight galaxies previously known to have emission lines, the narrow line region can be decomposed in two components with distinct line widths. In addition to this, broad H$\alpha$ emission was detected in six galaxies. The two galaxies not previously known to have emission lines show weak H$\alpha$+[N II] lines. All 10 galaxies may be classified as low-ionization nuclear emission regions in diagnostic diagrams and seven of them have bona fide active galactic nuclei with luminosities between 10$^{40}$ and 10$^{43}$ erg s$^{-1}$. Eddington ratios are always < 10$^{-3}$.

NGC 7097: THE ACTIVE GALACTIC NUCLEUS AND ITS MIRROR, REVEALED BY PRINCIPAL COMPONENT ANALYSIS TOMOGRAPHY

Three-dimensional (3D) spectroscopy techniques are becoming more and more popular, producing an increasing number of large data cubes. The challenge of extracting information from these cubes requires the development of new techniques for data processing and analysis. We apply the recently developed technique of Principal Component Analysis (PCA) Tomography to a data cube from the center of the elliptical galaxy NGC 7097 and show that this technique is effective in decomposing the data into physically interpretable information. We find that the first five principal components of our data are associated with distinct physical characteristics. In particular, we detect a LINER with a weak broad component in the Balmer lines. Two images of the LINER are present in our data, one seen through a disk of gas and dust, and the other after scattering by free electrons and/or dust particles in the ionization cone. Furthermore, we extract the spectrum of the LINER, decontaminated from stellar and extended nebular emission, using only the technique of PCA Tomography. We anticipate that the scattered image has polarized light, due to its scattered nature.

The Active Nucleus in NGC 4579

In this work, we present an analysis of a data cube obtained with the instrument IFU/GMOS Gemini North telescope centered on the nuclear region of the LINER galaxy NGC 4579. This galaxy is known to have a type 1 AGN (see Eracleous et al. 2002 for a review). The methodology used for the analysis of the data cube was PCA tomography (Steiner et al. 2009), which consists of applying the statistical tool known as principal component analysis (PCA) to extract information from data cubes. The analysis was performed in the following way: first, the data cube for this object was transformed into a data matrix, where each row corresponded to a spatial pixel and each column corresponded to a spectral pixel. We then applied the PCA to this matrix, obtaining eigenvectors as a function of spectral pixels (which correspond to the original coordinates). These eigenvectors, which have the appearance of spectra, are called eigenspectra. We also calculated the projection of the data corresponding to the spatial pixels on each one of the eigenvectors. These images were called tomograms. Figure 1 shows Eigenspectrum E2 and tomogram T2 superposed on T1.

PCA Tomography: how to extract information from data cubes

Astronomy has evolved almost exclusively by the use of spectroscopic and imaging techniques, operated separately. With the development of modern technologies, it is possible to obtain data cubes in which one combines both techniques simultaneously, producing images with spectral resolution. To extract information from them can be quite complex, and hence the development of new methods of data analysis is desirable. We present a method of analysis of data cube (data from single field observations, containing two spatial and one spectral dimension) that uses Principal Component Analysis (PCA) to express the data in the form of reduced dimensionality, facilitating efficient information extraction from very large data sets. PCA transforms the system of correlated coordinates into a system of uncorrelated coordinates ordered by principal components of decreasing variance. The new coordinates are referred to as eigenvectors, and the projections of the data on to these coordinates produce images we will call tomograms. The association of the tomograms (images) to eigenvectors (spectra) is important for the interpretation of both. The eigenvectors are mutually orthogonal, and this information is fundamental for their handling and interpretation. When the data cube shows objects that present uncorrelated physical phenomena, the eigenvector's orthogonality may be instrumental in separating and identifying them. By handling eigenvectors and tomograms, one can enhance features, extract noise, compress data, extract spectra, etc. We applied the method, for illustration purpose only, to the central region of the low ionization nuclear emission region (LINER) galaxy NGC 4736, and demonstrate that it has a type 1 active nucleus, not known before. Furthermore, we show that it is displaced from the centre of its stellar bulge.