Predicting Resolved Dust Attenuation from Local Galaxy Properties Using MaNGA

Abstract

Accurate spatially resolved dust corrections are critical for interpreting the structure and evolution of star-forming galaxies (SFGs). We present an empirical model for predicting spatially resolved dust attenuation ($A_V$) in SFGs using integral field spectroscopy from the Mapping Nearby Galaxies at Apache Point Observatory (MaNGA) survey. Using a sample of 5,155 galaxies over $7.20<M_\ast<11.14$ and $0.0002 < z < 0.1444$, we derive $A_V$ maps from the Balmer decrement across more than 1,898,954 star-forming spaxels. Using local star formation rate surface density ($Σ_{\text{SFR}}$) as a predictor, the model achieves $R^2 = 0.69$ and RMSE $=0.22$ mag, with residuals that are approximately Gaussian and centred near zero. It predicts $A_V$ within a factor of $\sim$1.3 on kpc scales. We also demonstrate that the relation can be applied iteratively to recover dust-corrected $Σ_{\mathrm{SFR}}$ from uncorrected values, converging by the fourth iteration with minimal residual bias ($-0.01$ mag) and low RMSE ($0.42$ mag). The model accurately reproduces $A_V$ maps across diverse morphologies and orientations, including edge-on systems. It also recovers the observed radial $A_V$ profiles, capturing their dependence on stellar mass and relative star formation activity, with more massive and more strongly star-forming galaxies showing steeper gradients.

Figures (9)

• Figure 1: Corner plot showing the distributions and intrinsic relations between $\log_{10}\,\Sigma_\ast$, $\log_{10}\,\Sigma_{\mathrm{SFR}}$, $R/R_e$, and $A_V$ for star-forming spaxels in our sample. The diagonal panels display one-dimensional histograms with vertical lines marking the 16th, 50th, and 84th percentiles. The lower–triangle panels show the corresponding distributions with contours enclosing 25%, 50%, and 90% of the data, with Spearman correlation coefficients $r$ indicated in each panel.
• Figure 2: Residual distributions ($A_V - A_{V,\mathrm{pred}}$) for different OLS linear models. The x-axis shows residuals between observed and predicted $A_V$, and the y-axis shows the number of spaxels per bin. Each model, based on different combinations of predictors, is colour-coded.
• Figure 3: Distribution of $A_V$ over the $\log \Sigma_{\mathrm{SFR}}$-$\log \Sigma_\ast$ plane. Panels (a) and (b) show the observed and predicted (see Equation \ref{['eq:linear_av_from_sigma_sfr']}) median $A_V$ values per bin (0.05 dex), respectively. Panels (c) and (d) display the median residuals ($A_V^\mathrm{obs} - A_V^\mathrm{pred}$) and associated standard deviation, $\sigma\left(A_V^\mathrm{obs} - A_V^\mathrm{pred}\right)$, respectively. Black contours enclose 90% and 50% of the sample.
• Figure 4: Residuals of the predicted $A_V$ from the empirical model as a function of (from left to right) observed $A_V$, $\log_{10} \Sigma_\ast$, $\log_{10} (\Sigma_{\mathrm{SFR}})$, and normalised galactocentric radius ($R/R_e$). In each panel, colours indicate the median $\log_{10}$ H$\beta$ SNR in each 2D bin, with 50% and 90% spaxel density contours overlaid in black. Horizontal dashed lines mark zero residual. The rightmost panel shows the overall residual distribution as a histogram.
• Figure 5: Residuals of the predicted $A_V$ from the iterative empirical correction. From left to right: (1) distributions of residuals for the first five iterations, (2) RMSE of the residuals as a function of iteration number, and (3) median residual offset as a function of iteration. Dashed vertical and horizontal lines indicate zero residual.