The MASSIVE SURVEY XXI: Local Variations in the Stellar Initial Mass Function of MASSIVE Early-Type Galaxies

Abstract

Extensive evidence suggests that the stellar initial mass function (IMF) varies among early-type galaxies (ETGs), but spatially resolved studies within individual galaxies are limited in sample size. We investigate radial variations in the low-mass ($\le 1M_{\odot }$) IMF and its connection to stellar populations in 37 nearby massive ETGs from the MASSIVE survey. Using high-quality Magellan/LDSS-3 long-slit spectroscopy spanning $0.4\hat{I}¼$m$-1.01\hat{I}¼$m, we extract spectra in radial bins reaching outermost radii of 0.2-1.1Re across the sample. We find that the IMF becomes less bottom-heavy with increasing radius in most galaxies. The sample-averaged IMF mismatch parameter, $\hat{I}{\pm }_{\mathrm{IMF}}=(M/L)/(M/L{)}_{\mathrm{Kroupa}}$, decreases from 2.16 within Re/8 to 1.74 in the Re/4-Re/2 bin, with galaxy-to-galaxy scatters of 0.50 and 0.42, respectively. Thus, the average IMF remains more bottom-heavy than Kroupa and approximately Salpeter-like or more bottom-heavy over these radii. The radial gradients of $\log (\hat{I}{\pm }_{\mathrm{IMF}})$ anti-correlate with the central value of $\hat{I}{\pm }_{\mathrm{IMF}}$, indicating that galaxies with more bottom-heavy central IMFs decline more steeply toward less bottom-heavy, approximately Salpeter-like values at larger radii. We find mild positive local correlations between $\hat{I}{\pm }_{\mathrm{IMF}}$ and stellar metallicity, but no significant local correlation with [Mg/Fe] or [Na/Fe]. Together with the approximately flat profiles of several [$\hat{I}\pm$/Fe], this suggests that IMF variation in massive ETGs is more closely linked to metallicity than to the star-formation timescale traced by [$\hat{I}\pm$/Fe]. Finally, the radial variation in stellar $M/L_r$ is dominated by the IMF gradient rather than by the stellar-population gradient. A fixed Kroupa IMF underestimates stellar masses by factors of 1.7 and 1.5 within Re/2 and Re in massive ETGs.

23 pages, 10 figures, submitted to ApJ