Radiative and mechanical feedback into the molecular gas of NGC 253

¹ Sterrewacht Leiden, Universiteit Leiden, PO Box 9513, 2300 RA Leiden, The Netherlands
e-mail: rosenberg@strw.leidenuniv.nl
² Kapteyn Astronomical Institute, University of Groningen, PO Box 800, 9700 AV Groningen, The Netherlands
³ Max-Planck-Institut für Radioastronomie, auf dem Hügel 16, Bonn 53121, Germany

Received: 22 November 2013
Accepted: 20 January 2014

Abstract

Starburst galaxies are galaxies or regions of galaxies undergoing intense periods of star formation. Understanding the heating and cooling mechanisms in these galaxies can give us insight to the driving mechanisms that fuel the starburst. Molecular emission lines play a crucial role in the cooling of the excited gas. With Herschel Spectral and Photometric Imaging Receiver we have been able to observe the rich molecular spectrum towards the central region of NGC 253. Carbon monoxide (CO, J = 4 − 3 to 13−12) is the brightest molecule in the Herschel wavelength range and together with ground-based low-J observations, the line fluxes trace the excitation of CO. By studying the CO excitation ladder and comparing the intensities to models, we investigate whether the gas is excited by UV radiation, X-rays, cosmic rays, or turbulent heating. Comparing the ¹²CO and ¹³CO observations to large velocity gradient models and photon-dominated region (PDR) models we find three main interstellar medium (ISM) phases. We estimate the density, temperature, and masses of these ISM phases. By adding ¹³CO, HCN, and HNC line intensities, we are able to constrain these degeneracies and determine the heating sources. The first ISM phase responsible for the low-J CO lines is excited by PDRs, but the second and third phases, responsible for the mid to high-J CO transitions, require an additional heating source. We find three possible combinations of models that can reproduce our observed molecular emission. Although we cannot determine which of these is preferable, we can conclude that mechanical heating is necessary to reproduce the observed molecular emission and cosmic ray heating is a negligible heating source. We then estimate the mass of each ISM phase; 6 × 10⁷M_⊙ for phase 1 (low-J CO lines), 3 × 10⁷M_⊙ for phase 2 (mid-J CO lines), and 9 × 10⁶M_⊙ for phase 3 (high-J CO lines) for a total system mass of 1 × 10⁸M_⊙.