The stellar mass-to-light ratio gradient in SDSS $r-$band $ abla (M_*/L_r)$ of a galaxy depends on its mass assembly history, which is imprinted in its morphology and gradients of age, metallicity, and stellar initial mass function (IMF). Taking a MaNGA sample of 2051 galaxies with stellar masses ranging from $10^9$ to $10^{12}M_odot$ released in SDSS DR15, we focus on face-on galaxies, without merger and bar signatures, and investigate the dependence of the 2D $ abla (M_*/L_r)$ on other galaxy properties, including $M_*/L_r$-colour relationships by assuming a fixed Salpeter IMF as the mass normalization reference. The median gradient is $ abla M_*/L_rsim -0.1$ (i.e., the $M_*/L_r$ is larger at the centre) for massive galaxies, becomes flat around $M_*sim 10^{10} M_{odot}$ and change sign to $ abla M_*/L_rsim 0.1$ at the lowest masses. The $M_*/L_r$ inside a half light radius increases with increasing galaxy stellar mass; in each mass bin, early-type galaxies have the highest value, while pure-disk late-type galaxies have the smallest. Correlation analyses suggest that the mass-weighted stellar age is the dominant parameter influencing the $M_*/L_r$ profile, since a luminosity-weighted age is easily affected by star formation when the specific star formation rate (sSFR) inside the half light radius is higher than $10^{-3} {rm Gyr}^{-1}$. With increased sSFR gradient, one can obtain a steeper negative $ abla (M_*/L_r)$. The scatter in the slopes of $M_*/L$-colour relations increases with increasing sSFR, for example, the slope for post-starburst galaxies can be flattened to $0.45$ from the global value $0.87$ in the $M_*/L$ vs. $g-r$ diagram. Hence converting galaxy colours to $M_*/L$ should be done carefully, especially for those galaxies with young luminosity-weighted stellar ages, which can have quite different star formation histories.
Download